Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

chat_template.jinja

@@ -0,0 +1,4 @@
+{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '
+' + message['content'] | trim + '<end_of_turn>
+' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model
+'}}{% endif %}

config.json

@@ -0,0 +1,40 @@
+{
+  "architectures": [
+    "Gemma2MoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_soft_capping": 50.0,
+  "auto_map": {
+    "AutoConfig": "suayptalha/turkish-gemmoe2-t1-base--configuration_gemma2moe.Gemma2MoeConfig",
+    "AutoModelForCausalLM": "suayptalha/turkish-gemmoe2-t1-base--modeling_gemma2moe.Gemma2MoeForCausalLM"
+  },
+  "bos_token_id": 2,
+  "dtype": "bfloat16",
+  "eos_token_id": 1,
+  "head_dim": 256,
+  "hidden_act": "gelu_pytorch_tanh",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "logit_soft_capping": 30.0,
+  "max_position_embeddings": 8192,
+  "model_type": "gemma2moe",
+  "num_attention_heads": 16,
+  "num_experts_per_tok": 2,
+  "num_hidden_layers": 42,
+  "num_key_value_heads": 8,
+  "num_local_experts": 3,
+  "output_router_logits": false,
+  "pad_token_id": 0,
+  "path": "suayptalha/turkish-gemmoe2-t1-base",
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "router_aux_loss_coef": 0.001,
+  "router_jitter_noise": 0.0,
+  "sliding_window": 4096,
+  "transformers_version": "4.56.0",
+  "use_cache": true,
+  "vocab_size": 256000
+}

configuration_gemma2moe.py

@@ -0,0 +1,110 @@
+# coding=utf-8
+# Copyright 2024 The HuggingFace Inc. team and Gemma2MoE Contributors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Gemma2MoE model configuration"""
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+class Gemma2MoeConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`Gemma2MoeModel`]. It is used to instantiate an Gemma2MoE
+    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+    defaults will yield a similar configuration to that of the Gemma-2-9b but with MoE capabilities.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+    """
+    model_type = "gemma2moe"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=256000,
+        hidden_size=3584,
+        intermediate_size=14336,
+        num_hidden_layers=42,
+        num_attention_heads=16,
+        num_key_value_heads=8,
+        head_dim=256,
+        hidden_act="gelu_pytorch_tanh",
+        max_position_embeddings=8192,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        pad_token_id=0,
+        eos_token_id=1,
+        bos_token_id=2,
+        tie_word_embeddings=True,
+        rope_theta=10000.0,
+        attention_bias=False,
+        attention_dropout=0.0,
+        
+        # Gemma 2 Specific Args
+        query_pre_attn_scalar=224, # 1/sqrt(head_dim) yerine Gemma2'ye özel scaling (genelde hidden_size temelli)
+        sliding_window=4096,       # Sliding Window Attention window size
+        logit_soft_capping=30.0,   # Final logit soft capping
+        attn_logit_soft_capping=50.0, # Attention scores soft capping
+        
+        # MoE Arguments
+        num_experts_per_tok=2,
+        num_local_experts=8,
+        router_aux_loss_coef=0.001,
+        output_router_logits=False,
+        router_jitter_noise=0.0,   # Opsiyonel: Router stabilitesi için jitter
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        
+        # Grouped Query Attention (GQA) kontrolü
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+        self.num_key_value_heads = num_key_value_heads
+        
+        self.head_dim = head_dim
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+
+        # Gemma 2 Specifics
+        self.query_pre_attn_scalar = query_pre_attn_scalar
+        self.sliding_window = sliding_window
+        self.logit_soft_capping = logit_soft_capping
+        self.attn_logit_soft_capping = attn_logit_soft_capping
+
+        # MoE Specifics
+        self.num_experts_per_tok = num_experts_per_tok
+        self.num_local_experts = num_local_experts
+        self.router_aux_loss_coef = router_aux_loss_coef
+        self.output_router_logits = output_router_logits
+        self.router_jitter_noise = router_jitter_noise
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.56.0"
+}

model-00001-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1458267106e3e74e4271437e790699f480e47fa822429d5d5ea9aa0c7920f5a
+size 4962041280

model-00002-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aca5cdb17ec10599608bc342de0bd93a2d90e1ac5f64914079b7c1f2c4a3ddcb
+size 4976781976

model-00003-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef271ea9dfdeb64b15f0a0f7a10d4be0dda524838b9d9ece9fad0ab70e01e17a
+size 4962094760

model-00004-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1984189ecc983997d2e62317039b29ac3c91c6cfae5437e47c2084581603c51c
+size 4962123960

model-00005-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de0a692c6b4c4c3d7c3886141a3b0a80a4cccb224bb530c4d7d8a048c7b8d7c6
+size 4962123960

model-00006-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4eb0b93d14004a42572536b94f6e338705136a2230da60664f3f6b82a30d49ed
+size 4962123960

model-00007-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f89363a36e54457acf5072084e568575762b184e7365773cced52ab50b3ad8d
+size 4962123960

model-00008-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c497dd8b0f3057f51838011fcec114a1157443260f55025bcf354d27e1318d
+size 4962123960

model-00009-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:380abb7b99f0065222ac3d120c35c58230674aef25027274acd2a733d04ec6c3
+size 4668507560

model.safetensors.index.json

@@ -0,0 +1,766 @@
+{
+  "metadata": {
+    "total_parameters": 22189974016,
+    "total_size": 44379948032
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.0.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.0.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.0.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.1.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.1.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.1.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.2.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.2.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.experts.2.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.block_sparse_moe.gate.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.0.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.0.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.0.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.1.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.1.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.1.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.2.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.2.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.experts.2.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.block_sparse_moe.gate.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.0.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.0.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.0.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.1.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.1.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.1.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.2.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.2.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.experts.2.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.block_sparse_moe.gate.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.post_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.pre_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.0.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.0.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.0.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.1.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.1.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.1.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.2.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.2.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.experts.2.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.block_sparse_moe.gate.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.post_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.pre_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.0.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.0.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.0.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.1.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.1.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.1.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.2.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.2.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.experts.2.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.block_sparse_moe.gate.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.12.post_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.12.pre_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.0.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.0.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.0.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.1.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.1.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.1.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.2.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.2.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.experts.2.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.block_sparse_moe.gate.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.post_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.pre_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.0.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.0.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.0.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.1.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.1.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.1.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.2.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.2.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.experts.2.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.block_sparse_moe.gate.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.post_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.pre_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.0.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.0.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.0.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.1.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.1.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.1.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.2.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.2.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.experts.2.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.block_sparse_moe.gate.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.post_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.pre_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.0.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.0.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.0.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.1.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.1.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.1.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.2.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.2.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.experts.2.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.block_sparse_moe.gate.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.post_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.pre_feedforward_layernorm.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.0.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.0.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.0.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.1.down_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.1.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.1.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.2.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.2.gate_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.experts.2.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.17.block_sparse_moe.gate.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.17.post_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.17.pre_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.0.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.0.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.0.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.1.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.1.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.1.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.2.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.2.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.experts.2.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.block_sparse_moe.gate.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.post_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.pre_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.0.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.0.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.0.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.1.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.1.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.1.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.2.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.2.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.experts.2.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.block_sparse_moe.gate.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.post_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.pre_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.0.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.0.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.0.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.1.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.1.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.1.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.2.down_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.2.gate_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.experts.2.up_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.block_sparse_moe.gate.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.0.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.0.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.0.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.1.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.1.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.1.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.2.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.2.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.experts.2.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.block_sparse_moe.gate.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.post_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.pre_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.0.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.0.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.0.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.1.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.1.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.1.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.2.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.2.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.experts.2.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.block_sparse_moe.gate.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.post_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.pre_feedforward_layernorm.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.0.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.0.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.0.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.1.down_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.1.gate_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.1.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.2.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.2.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.experts.2.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.block_sparse_moe.gate.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.post_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.pre_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.0.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.0.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.0.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.1.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.1.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.1.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.2.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.2.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.experts.2.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.block_sparse_moe.gate.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.post_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.pre_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.0.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.0.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.0.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.1.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.1.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.1.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.2.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.2.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.experts.2.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.block_sparse_moe.gate.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.post_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.pre_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.0.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.0.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.0.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.1.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.1.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.1.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.2.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.2.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.experts.2.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.block_sparse_moe.gate.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.post_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.pre_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.0.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.0.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.0.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.1.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.1.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.1.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.2.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.2.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.experts.2.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.block_sparse_moe.gate.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.post_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.pre_feedforward_layernorm.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.0.down_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.0.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.0.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.1.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.1.gate_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.1.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.2.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.2.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.experts.2.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.block_sparse_moe.gate.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.post_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.pre_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.0.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.0.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.0.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.1.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.1.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.1.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.2.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.2.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.experts.2.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.block_sparse_moe.gate.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.post_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.pre_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.0.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.0.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.0.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.1.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.1.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.1.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.2.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.2.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.experts.2.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.block_sparse_moe.gate.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.post_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.pre_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.0.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.0.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.0.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.1.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.1.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.1.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.2.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.2.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.experts.2.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.block_sparse_moe.gate.weight": "model-00001-of-00009.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.0.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.0.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.0.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.1.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.1.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.1.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.2.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.2.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.experts.2.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.block_sparse_moe.gate.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.post_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.pre_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.0.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.0.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.0.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.1.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.1.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.1.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.2.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.2.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.experts.2.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.block_sparse_moe.gate.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.post_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.pre_feedforward_layernorm.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.0.down_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.0.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.0.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.1.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.1.gate_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.1.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.2.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.2.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.experts.2.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.block_sparse_moe.gate.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.post_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.pre_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.0.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.0.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.0.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.1.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.1.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.1.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.2.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.2.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.experts.2.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.block_sparse_moe.gate.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.post_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.pre_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.0.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.0.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.0.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.1.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.1.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.1.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.2.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.2.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.experts.2.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.block_sparse_moe.gate.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.post_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.pre_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.0.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.0.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.0.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.1.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.1.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.1.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.2.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.2.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.experts.2.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.block_sparse_moe.gate.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.post_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.pre_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.0.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.0.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.0.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.1.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.1.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.1.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.2.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.2.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.experts.2.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.block_sparse_moe.gate.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.post_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.pre_feedforward_layernorm.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.0.down_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.0.gate_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.0.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.1.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.1.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.1.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.2.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.2.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.experts.2.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.block_sparse_moe.gate.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.post_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.pre_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.0.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.0.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.0.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.1.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.1.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.1.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.2.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.2.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.experts.2.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.block_sparse_moe.gate.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.post_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.pre_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.0.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.0.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.0.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.1.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.1.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.1.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.2.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.2.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.experts.2.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.block_sparse_moe.gate.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.post_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.pre_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.0.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.0.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.0.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.1.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.1.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.1.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.2.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.2.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.experts.2.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.block_sparse_moe.gate.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.0.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.0.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.0.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.1.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.1.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.1.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.2.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.2.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.experts.2.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.block_sparse_moe.gate.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.post_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.pre_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.0.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.0.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.0.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.1.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.1.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.1.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.2.down_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.2.gate_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.experts.2.up_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.block_sparse_moe.gate.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.post_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.pre_feedforward_layernorm.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00009-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.0.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.0.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.0.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.1.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.1.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.1.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.2.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.2.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.experts.2.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.block_sparse_moe.gate.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.0.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.0.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.0.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.1.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.1.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.1.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.2.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.2.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.experts.2.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.block_sparse_moe.gate.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.0.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.0.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.0.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.1.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.1.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.1.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.2.down_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.2.gate_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.experts.2.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.block_sparse_moe.gate.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.0.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.0.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.0.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.1.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.1.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.1.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.2.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.2.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.experts.2.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.block_sparse_moe.gate.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.post_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.pre_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.0.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.0.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.0.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.1.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.1.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.1.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.2.down_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.2.gate_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.experts.2.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.block_sparse_moe.gate.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.post_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.pre_feedforward_layernorm.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00009.safetensors",
+    "model.norm.weight": "model-00009-of-00009.safetensors"
+  }
+}

modeling_gemma2moe.py

@@ -0,0 +1,753 @@
+# coding=utf-8
+# Copyright 2024 The HuggingFace Inc. team and Gemma2MoE Contributors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch Gemma2MoE model."""
+
+import math
+from typing import List, Optional, Tuple, Union
+
+import torch
+import torch.nn.functional as F
+import torch.utils.checkpoint
+from torch import nn
+from torch.nn import CrossEntropyLoss
+
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache, StaticCache
+from transformers.modeling_outputs import MoeModelOutputWithPast, MoeCausalLMOutputWithPast
+from transformers.modeling_utils import PreTrainedModel
+from transformers.generation import GenerationMixin
+from transformers.utils import (
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    logging,
+    replace_return_docstrings,
+)
+from transformers.modeling_attn_mask_utils import _prepare_4d_causal_attention_mask
+
+from .configuration_gemma2moe import Gemma2MoeConfig
+
+logger = logging.get_logger(__name__)
+
+_CONFIG_FOR_DOC = "Gemma2MoeConfig"
+
+# --- Auxiliary Loss & Router Functions ---
+
+def load_balancing_loss_func(gate_logits: torch.Tensor, num_experts: torch.Tensor = None, top_k=2) -> float:
+    r"""
+    Computes auxiliary load balancing loss as in Switch Transformer.
+    """
+    if gate_logits is None or not isinstance(gate_logits, torch.Tensor):
+        return 0.0
+
+    # gate_logits: [batch_size * seq_len, num_experts] assumed flattened or [batch, seq, experts]
+    if gate_logits.dim() == 3:
+        gate_logits = gate_logits.view(-1, gate_logits.shape[-1])
+
+    routing_weights = torch.softmax(gate_logits, dim=-1)
+    
+    # top_k indices
+    _, selected_experts = torch.topk(routing_weights, top_k, dim=-1)
+
+    # expert_mask: [num_tokens, num_experts]
+    expert_mask = torch.nn.functional.one_hot(selected_experts, num_experts)
+    if expert_mask.dim() == 3:
+        expert_mask = expert_mask.sum(dim=1) # Sum over k selected experts
+    
+    # Normalize to get fraction of tokens per expert
+    tokens_per_expert = torch.mean(expert_mask.float(), dim=0)
+    
+    # Mean probability per expert
+    router_prob_per_expert = torch.mean(routing_weights, dim=0)
+
+    # Loss = N * sum(f_i * P_i)
+    overall_loss = torch.sum(tokens_per_expert * router_prob_per_expert) * num_experts
+    return overall_loss
+
+
+# --- Gemma 2 Components ---
+
+class Gemma2RMSNorm(nn.Module):
+    def __init__(self, dim: int, eps: float = 1e-6):
+        super().__init__()
+        self.eps = eps
+        self.weight = nn.Parameter(torch.zeros(dim))
+
+    def _norm(self, x):
+        return x * torch.rsqrt(x.pow(2).mean(-1, keepdim=True) + self.eps)
+
+    def forward(self, x):
+        output = self._norm(x.float())
+        # Gemma 2 signature: output * (1 + weight)
+        # Casting back to input dtype
+        output = output * (1.0 + self.weight.float())
+        return output.type_as(x)
+
+
+class Gemma2RotaryEmbedding(nn.Module):
+    def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None):
+        super().__init__()
+        self.dim = dim
+        self.max_position_embeddings = max_position_embeddings
+        self.base = base
+        self.register_buffer("inv_freq", None, persistent=False)
+
+    @torch.no_grad()
+    def forward(self, x, position_ids, seq_len=None):
+        # x: [bs, num_attention_heads, seq_len, head_size]
+        if self.inv_freq is None:
+            self.inv_freq = 1.0 / (
+                self.base ** (torch.arange(0, self.dim, 2, dtype=torch.int64, device=x.device).float() / self.dim)
+            )
+        inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1)
+        position_ids_expanded = position_ids[:, None, :].float()
+        
+        # Use float32 for RoPE calculation to maintain precision
+        with torch.autocast(device_type=x.device.type, enabled=False):
+            freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2)
+            emb = torch.cat((freqs, freqs), dim=-1)
+            cos = emb.cos()
+            sin = emb.sin()
+        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+
+def apply_rotary_pos_emb(q, k, cos, sin):
+    """Applies Rotary Position Embedding to the query and key tensors."""
+    cos = cos.unsqueeze(1) # [bs, 1, seq_len, head_dim]
+    sin = sin.unsqueeze(1)
+    q_embed = (q * cos) + (rotate_half(q) * sin)
+    k_embed = (k * cos) + (rotate_half(k) * sin)
+    return q_embed, k_embed
+
+
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep).
+    Used for Grouped Query Attention (GQA).
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+class Gemma2Attention(nn.Module):
+    """
+    Multi-headed attention with Soft-capping, Sliding Window and GQA.
+    """
+    def __init__(self, config: Gemma2MoeConfig, layer_idx: Optional[int] = None):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        
+        self.attention_dropout = config.attention_dropout
+        self.hidden_size = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = config.head_dim
+        self.num_key_value_heads = config.num_key_value_heads
+        self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.max_position_embeddings = config.max_position_embeddings
+        self.rope_theta = config.rope_theta
+        self.is_causal = True
+        
+        # Gemma 2 scaling specific
+        self.scaling = config.query_pre_attn_scalar ** -0.5
+        
+        # Soft capping parameter
+        self.attn_logit_soft_capping = config.attn_logit_soft_capping
+
+        self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.attention_bias)
+        self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=config.attention_bias)
+        self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=config.attention_bias)
+        self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=config.attention_bias)
+        
+        self.rotary_emb = Gemma2RotaryEmbedding(
+            self.head_dim,
+            max_position_embeddings=self.max_position_embeddings,
+            base=self.rope_theta,
+        )
+        self.sliding_window = config.sliding_window
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        cache_position: Optional[torch.LongTensor] = None,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        bsz, q_len, _ = hidden_states.size()
+
+        query_states = self.q_proj(hidden_states)
+        key_states = self.k_proj(hidden_states)
+        value_states = self.v_proj(hidden_states)
+
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        cos, sin = self.rotary_emb(value_states, position_ids=position_ids, seq_len=None)
+        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
+
+        if past_key_value is not None:
+             # cache_position for static cache, legacy for dynamic
+            cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position, "sliding_window": self.sliding_window}
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        key_states = repeat_kv(key_states, self.num_key_value_groups)
+        value_states = repeat_kv(value_states, self.num_key_value_groups)
+
+        # Scaled Dot Product Calculation
+        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) * self.scaling
+
+        # Logit Soft Capping
+        if self.attn_logit_soft_capping is not None:
+            attn_weights = attn_weights / self.attn_logit_soft_capping
+            attn_weights = torch.tanh(attn_weights)
+            attn_weights = attn_weights * self.attn_logit_soft_capping
+
+        if attention_mask is not None:
+            # Mask should be broadcastable
+            attn_weights = attn_weights + attention_mask
+
+        # Softmax and Dropout
+        attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype)
+        attn_weights = nn.functional.dropout(attn_weights, p=self.attention_dropout, training=self.training)
+        
+        attn_output = torch.matmul(attn_weights, value_states)
+
+        if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim):
+            raise ValueError(
+                f"`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is"
+                f" {attn_output.size()}"
+            )
+
+        attn_output = attn_output.transpose(1, 2).contiguous()
+        attn_output = attn_output.view(bsz, q_len, -1)
+        attn_output = self.o_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+
+        return attn_output, attn_weights, past_key_value
+
+
+# --- Expert & MoE Block ---
+
+class Gemma2MLP(nn.Module):
+    """
+    Gemma 2 MLP: Gated GELU Tanh
+    """
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = config.intermediate_size
+        
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.act_fn = ACT2FN[config.hidden_act]
+
+    def forward(self, x):
+        return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+
+
+class Gemma2MoeBlock(nn.Module):
+    """
+    Sparse MoE Block for Gemma 2.
+    Uses Top-k gating and processes selected tokens through experts.
+    """
+    def __init__(self, config: Gemma2MoeConfig):
+        super().__init__()
+        self.hidden_dim = config.hidden_size
+        self.num_experts = config.num_local_experts
+        self.top_k = config.num_experts_per_tok
+        self.jitter_noise = config.router_jitter_noise
+
+        self.gate = nn.Linear(self.hidden_dim, self.num_experts, bias=False)
+        self.experts = nn.ModuleList([Gemma2MLP(config) for _ in range(self.num_experts)])
+
+    def forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]:
+        batch_size, sequence_length, hidden_dim = hidden_states.shape
+        hidden_states_flat = hidden_states.view(-1, hidden_dim)
+
+        # Router Logits
+        router_logits = self.gate(hidden_states_flat)
+        
+        if self.training and self.jitter_noise > 0:
+            router_logits += torch.empty_like(router_logits).uniform_(1.0 - self.jitter_noise, 1.0 + self.jitter_noise)
+
+        routing_weights = F.softmax(router_logits, dim=1)
+        topk_weight, topk_idx = torch.topk(routing_weights, self.top_k, dim=-1, sorted=False)
+        
+        # Normalize weights
+        topk_weight /= topk_weight.sum(dim=-1, keepdim=True)
+        topk_weight = topk_weight.to(hidden_states.dtype)
+
+        # Routing process
+        # Using a loop here for clarity and simplicity in Python. 
+        # For extreme performance, Triton or CUDA kernels should be used.
+        final_hidden_states = torch.zeros_like(hidden_states_flat)
+        
+        # Flatten indices to handle batching easier
+        flat_topk_idx = topk_idx.view(-1) 
+        
+        # We need to process each expert
+        for i, expert in enumerate(self.experts):
+            # Find tokens assigned to this expert (in any of the top-k slots)
+            # This is a bit inefficient in pure PyTorch but ensures correctness without custom kernels
+            # Create a mask for tokens where this expert is selected
+            expert_mask = (topk_idx == i)
+            
+            if expert_mask.any():
+                # We need to collect inputs, process, and scatter back
+                # This logic handles cases where an expert is selected multiple times (unlikely in top-k but possible conceptually)
+                # But typically top-k implies distinct experts.
+                
+                # Get indices where this expert is used
+                batch_indices, k_indices = torch.where(expert_mask)
+                
+                # Extract inputs
+                inp = hidden_states_flat[batch_indices]
+                
+                # Forward pass
+                out = expert(inp)
+                
+                # Weighting: We need the weight associated with this selection
+                weights = topk_weight[batch_indices, k_indices]
+                
+                # Accumulate result
+                # Ideally, scatter_add, but here we iterate. 
+                # Since batch_indices might repeat if we allowed k repetitions (we don't usually),
+                # standard scatter_add_ is safer.
+                
+                weighted_out = out * weights.unsqueeze(-1)
+                final_hidden_states.index_add_(0, batch_indices, weighted_out)
+
+        final_hidden_states = final_hidden_states.view(batch_size, sequence_length, hidden_dim)
+        return final_hidden_states, router_logits
+
+
+# --- Decoder Layer (Strict Gemma 2 Topology) ---
+
+class Gemma2MoeDecoderLayer(nn.Module):
+    def __init__(self, config: Gemma2MoeConfig, layer_idx: int):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+
+        self.self_attn = Gemma2Attention(config, layer_idx)
+        self.block_sparse_moe = Gemma2MoeBlock(config)
+        
+        # Gemma 2 uses 4 specific RMSNorms per layer
+        self.input_layernorm = Gemma2RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.post_attention_layernorm = Gemma2RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.pre_feedforward_layernorm = Gemma2RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.post_feedforward_layernorm = Gemma2RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Tuple[torch.Tensor]] = None,
+        output_attentions: Optional[bool] = False,
+        output_router_logits: Optional[bool] = False,
+        use_cache: Optional[bool] = False,
+        cache_position: Optional[torch.LongTensor] = None,
+    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
+        
+        # --- Attention Path ---
+        residual = hidden_states
+        hidden_states = self.input_layernorm(hidden_states)
+        
+        hidden_states, self_attn_weights, present_key_value = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_value=past_key_value,
+            output_attentions=output_attentions,
+            use_cache=use_cache,
+            cache_position=cache_position,
+        )
+        
+        hidden_states = self.post_attention_layernorm(hidden_states)
+        hidden_states = residual + hidden_states # Residual Connection
+
+        # --- MoE Path ---
+        residual = hidden_states
+        hidden_states = self.pre_feedforward_layernorm(hidden_states)
+        
+        # Using MoE instead of standard MLP
+        hidden_states, router_logits = self.block_sparse_moe(hidden_states)
+        
+        hidden_states = self.post_feedforward_layernorm(hidden_states)
+        hidden_states = residual + hidden_states # Residual Connection
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (self_attn_weights,)
+        
+        if use_cache:
+            outputs += (present_key_value,)
+            
+        if output_router_logits:
+            outputs += (router_logits,)
+
+        return outputs
+
+
+# --- PreTrained Model Wrappers ---
+
+class Gemma2MoePreTrainedModel(PreTrainedModel):
+    config_class = Gemma2MoeConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["Gemma2MoeDecoderLayer"]
+    _skip_keys_device_placement = ["past_key_values"]
+    _supports_flash_attn_2 = False # Keeping SDPA for broad compatibility logic implemented above
+    _supports_sdpa = True
+    _supports_cache_class = True
+
+    def _init_weights(self, module):
+        std = self.config.initializer_range
+        if isinstance(module, nn.Linear):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+
+
+class Gemma2MoeModel(Gemma2MoePreTrainedModel):
+    def __init__(self, config: Gemma2MoeConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [Gemma2MoeDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
+        )
+        self.norm = Gemma2RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.gradient_checkpointing = False
+        
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.embed_tokens = value
+
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        output_router_logits: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+    ) -> Union[Tuple, MoeModelOutputWithPast]:
+        
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_router_logits = output_router_logits if output_router_logits is not None else self.config.output_router_logits
+        output_hidden_states = output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embed_tokens(input_ids)
+
+        if use_cache and past_key_values is None:
+            past_key_values = DynamicCache()
+
+        if cache_position is None:
+            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            cache_position = torch.arange(
+                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+            )
+            
+        if position_ids is None:
+            position_ids = cache_position.unsqueeze(0)
+
+        # 4D Attention Mask Creation (handles Sliding Window if config requests it)
+        causal_mask = _prepare_4d_causal_attention_mask(
+            attention_mask,
+            (inputs_embeds.shape[0], inputs_embeds.shape[1]),
+            inputs_embeds,
+            past_key_values.get_seq_length() if past_key_values is not None else 0,
+            sliding_window=self.config.sliding_window,
+        )
+
+        # Normalization (Gemma 2 embedding scaling)
+        normalizer = torch.tensor(self.config.hidden_size**0.5, dtype=inputs_embeds.dtype)
+        hidden_states = inputs_embeds * normalizer
+
+        all_hidden_states = () if output_hidden_states else None
+        all_router_logits = () if output_router_logits else None
+        
+        for decoder_layer in self.layers:
+            if output_hidden_states:
+                all_hidden_states += (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+                layer_outputs = self._gradient_checkpointing_func(
+                    decoder_layer.__call__,
+                    hidden_states,
+                    causal_mask,
+                    position_ids,
+                    past_key_values,
+                    output_attentions,
+                    output_router_logits,
+                    use_cache,
+                    cache_position,
+                )
+            else:
+                layer_outputs = decoder_layer(
+                    hidden_states,
+                    attention_mask=causal_mask,
+                    position_ids=position_ids,
+                    past_key_value=past_key_values,
+                    output_attentions=output_attentions,
+                    output_router_logits=output_router_logits,
+                    use_cache=use_cache,
+                    cache_position=cache_position,
+                )
+
+            hidden_states = layer_outputs[0]
+
+            if output_router_logits:
+                all_router_logits += (layer_outputs[-1],)
+
+        hidden_states = self.norm(hidden_states)
+
+        if output_hidden_states:
+            all_hidden_states += (hidden_states,)
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, past_key_values, all_hidden_states, all_router_logits] if v is not None)
+
+        return MoeModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=past_key_values,
+            hidden_states=all_hidden_states,
+            router_logits=all_router_logits,
+        )
+
+
+class Gemma2MoeForCausalLM(Gemma2MoePreTrainedModel, GenerationMixin):
+    _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.model = Gemma2MoeModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+        self.router_aux_loss_coef = config.router_aux_loss_coef
+        self.num_experts = config.num_local_experts
+        self.num_experts_per_tok = config.num_experts_per_tok
+        
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        output_router_logits: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+    ) -> Union[Tuple, MoeCausalLMOutputWithPast]:
+        
+        output_router_logits = output_router_logits if output_router_logits is not None else self.config.output_router_logits
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            output_router_logits=output_router_logits,
+            return_dict=return_dict,
+            cache_position=cache_position,
+        )
+
+        hidden_states = outputs[0]
+        logits = self.lm_head(hidden_states)
+        
+        # Final Soft Capping (Gemma 2 Specific feature)
+        # tanh(logits / cap) * cap
+        if self.config.logit_soft_capping is not None:
+             logits = logits / self.config.logit_soft_capping
+             logits = torch.tanh(logits)
+             logits = logits * self.config.logit_soft_capping
+
+        logits = logits.float()
+
+        loss = None
+        if labels is not None:
+            shift_logits = logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            loss_fct = CrossEntropyLoss()
+            shift_logits = shift_logits.view(-1, self.config.vocab_size)
+            shift_labels = shift_labels.view(-1)
+            loss = loss_fct(shift_logits, shift_labels)
+
+        aux_loss = None
+        if output_router_logits:
+            aux_loss = load_balancing_loss_func(
+                outputs.router_logits if return_dict else outputs[-1],
+                self.num_experts,
+                self.num_experts_per_tok,
+            )
+            if labels is not None:
+                loss += self.router_aux_loss_coef * aux_loss
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            if output_router_logits:
+                output = (aux_loss,) + output
+            return (loss,) + output if loss is not None else output
+
+        return MoeCausalLMOutputWithPast(
+            loss=loss,
+            aux_loss=aux_loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            router_logits=outputs.router_logits,
+        )
+
+    def prepare_inputs_for_generation(
+        self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, **kwargs
+        ):
+        past_length = 0
+        if past_key_values is not None:
+            if isinstance(past_key_values, Cache):
+                past_length = cache_position[0] if cache_position is not None else past_key_values.get_seq_length()
+                
+                # --- HATA DÜZELTMESİ BAŞLANGICI ---
+                # get_max_length metodunun varlığını kontrol ediyoruz
+                if hasattr(past_key_values, "get_max_length") and past_key_values.get_max_length() is not None:
+                    max_cache_length = torch.tensor(past_key_values.get_max_length(), device=input_ids.device)
+                else:
+                    max_cache_length = None
+                # --- HATA DÜZELTMESİ BİTİŞİ ---
+
+                cache_length = past_length if max_cache_length is None else torch.min(max_cache_length, past_length)
+            
+            # Legacy Cache (Tuple formatı için)
+            else:
+                past_length = past_key_values[0][0].shape[2]
+                max_cache_length = None
+
+            # Keep only the unprocessed tokens:
+            if attention_mask is not None and attention_mask.shape[1] > input_ids.shape[1]:
+                input_ids = input_ids[:, -(attention_mask.shape[1] - past_length) :]
+            elif past_length < input_ids.shape[1]:
+                input_ids = input_ids[:, past_length:]
+
+            # If we are about to go beyond the maximum cache length, we need to crop the input attention mask.
+            if (
+                max_cache_length is not None
+                and attention_mask is not None
+                and cache_length + input_ids.shape[1] > max_cache_length
+            ):
+                attention_mask = attention_mask[:, -max_cache_length:]
+
+        position_ids = kwargs.get("position_ids", None)
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past_key_values:
+                position_ids = position_ids[:, -input_ids.shape[1] :]
+
+        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
+        if inputs_embeds is not None and past_key_values is None:
+            model_inputs = {"inputs_embeds": inputs_embeds}
+        else:
+            model_inputs = {"input_ids": input_ids}
+
+        if cache_position is None:
+            # Inputs embeds veya input_ids hangisi varsa onun shape'ini al
+            input_len = model_inputs.get("input_ids", inputs_embeds).shape[1]
+            cache_position = torch.arange(past_length, past_length + input_len, device=input_ids.device)
+        
+        model_inputs.update(
+            {
+                "position_ids": position_ids,
+                "cache_position": cache_position,
+                "past_key_values": past_key_values,
+                "use_cache": kwargs.get("use_cache"),
+                "attention_mask": attention_mask,
+            }
+        )
+        return model_inputs
+
+    @staticmethod
+    def _reorder_cache(past_key_values, beam_idx):
+        reordered_past = ()
+        for layer_past in past_key_values:
+            reordered_past += (
+                tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
+            )
+        return reordered_past

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a9dead7ec8422d57df8e480a53b45ea76e22cf7afc3d4b64587f8ea48aacbc8
+size 34363130

tokenizer_config.json

@@ -0,0 +1,2015 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<end_of_turn>",
+  "extra_special_tokens": {},
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizerFast",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}