Align model config with tokenizer settings

config.json

@@ -4,9 +4,9 @@
     "GPT2LMHeadModel"
   ],
   "attn_pdrop": 0.1,
-  "bos_token_id": 50256,
+  "bos_token_id": 2,
   "embd_pdrop": 0.1,
-  "eos_token_id": 50256,
+  "eos_token_id": 3,
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",
@@ -16,6 +16,7 @@
   "n_inner": null,
   "n_layer": 40,
   "n_positions": 2048,
+  "pad_token_id": 0,
   "reorder_and_upcast_attn": false,
   "resid_pdrop": 0.1,
   "scale_attn_by_inverse_layer_idx": false,

generation_config.json

@@ -1,6 +1,7 @@
 {
   "_from_model_config": true,
-  "bos_token_id": 50256,
-  "eos_token_id": 50256,
+  "bos_token_id": 2,
+  "eos_token_id": 3,
+  "pad_token_id": 0,
   "transformers_version": "4.27.1"
 }

tokenizer_config.json

@@ -20,7 +20,7 @@
   },
   "errors": "replace",
   "mask_token": "<mask>",
-  "model_max_length": 1000000000000000019884624838656,
+  "model_max_length": 2048,
   "pad_token": {
     "__type": "AddedToken",
     "content": "<pad>",
@@ -29,7 +29,9 @@
     "rstrip": false,
     "single_word": false
   },
+  "padding_side": "left",
   "tokenizer_class": "GPT2Tokenizer",
+  "truncation_side": "left",
   "unk_token": {
     "__type": "AddedToken",
     "content": "<|endoftext|>",