Upload folder using huggingface_hub

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "k_proj",
+    "q_proj",
     "o_proj",
+    "up_proj",
+    "v_proj",
     "down_proj",
-    "q_proj",
     "gate_proj",
-    "up_proj"
+    "k_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1ee538a62ccdced6ae2ff179651eb0c5481a976ce36e4d78f01b25a5a48351a
+oid sha256:95aa29e997e65895e7438c6a5161582325b48ad40f334444dd9e8bcf1af49f7b
 size 664584480

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4df917b3bdd630be19a2284b2cbf7e679189cc5cfce397f316886a3716f5ca3e
+oid sha256:9a18c2b2f06141f8d8ef924be72aecffd7eea284834e455f0e1231458243c1fd
 size 1329377575

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9da0163c3936acb3c6c189647b4d67e82c3ede0faa2bc4c2be5a6384e1d0721e
+oid sha256:a490ec3fd930ebc86317228df3023d06f035c382dedbf1c91b49dd6b00d113f2
 size 14645

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68e12ea105d1321d2f396e3113101572120c00e3b670a28d0154eb748559938c
+oid sha256:89c7e585bf25f929251599f24890acceb440d06b049a496384748f1eed552840
 size 1465

trainer_state.json

@@ -1,97 +1,118 @@
 {
-  "best_global_step": 80,
-  "best_metric": 0.2634693384170532,
-  "best_model_checkpoint": "/content/models/gemma_qlora_lmh_inst/checkpoint-80",
-  "epoch": 1.606060606060606,
+  "best_global_step": 100,
+  "best_metric": 0.0743941143155098,
+  "best_model_checkpoint": "/content/models/gemma_qlora_lmh_inst/checkpoint-100",
+  "epoch": 2.0,
   "eval_steps": 20,
-  "global_step": 80,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "entropy": 2.385788154602051,
+      "entropy": 2.200686830282211,
       "epoch": 0.40404040404040403,
-      "grad_norm": 1.4871121644973755,
+      "grad_norm": 2.258388042449951,
       "learning_rate": 8.1e-06,
-      "loss": 1.4621,
-      "mean_token_accuracy": 0.7171875,
-      "num_tokens": 234453.0,
+      "loss": 0.1306,
+      "mean_token_accuracy": 0.9350446417927742,
+      "num_tokens": 237653.0,
       "step": 20
     },
     {
       "epoch": 0.40404040404040403,
-      "eval_entropy": 2.444101461997399,
-      "eval_loss": 0.38041621446609497,
-      "eval_mean_token_accuracy": 0.7514423086093023,
-      "eval_num_tokens": 234453.0,
-      "eval_runtime": 5.7269,
-      "eval_samples_per_second": 34.399,
-      "eval_steps_per_second": 2.27,
+      "eval_entropy": 2.201928762289194,
+      "eval_loss": 0.0838593915104866,
+      "eval_mean_token_accuracy": 0.9620879109089191,
+      "eval_num_tokens": 237653.0,
+      "eval_runtime": 5.7826,
+      "eval_samples_per_second": 34.068,
+      "eval_steps_per_second": 2.248,
       "step": 20
     },
     {
-      "entropy": 2.449769389629364,
+      "entropy": 2.203989064693451,
       "epoch": 0.8080808080808081,
-      "grad_norm": 2.987805128097534,
+      "grad_norm": 3.0667307376861572,
       "learning_rate": 6.1e-06,
-      "loss": 0.3307,
-      "mean_token_accuracy": 0.8234375,
-      "num_tokens": 467222.0,
+      "loss": 0.0797,
+      "mean_token_accuracy": 0.9622767880558968,
+      "num_tokens": 473622.0,
       "step": 40
     },
     {
       "epoch": 0.8080808080808081,
-      "eval_entropy": 2.4303444899045505,
-      "eval_loss": 0.29822829365730286,
-      "eval_mean_token_accuracy": 0.8235576932246869,
-      "eval_num_tokens": 467222.0,
-      "eval_runtime": 5.7388,
-      "eval_samples_per_second": 34.328,
-      "eval_steps_per_second": 2.265,
+      "eval_entropy": 2.2022548638857327,
+      "eval_loss": 0.07604048401117325,
+      "eval_mean_token_accuracy": 0.9581043995343722,
+      "eval_num_tokens": 473622.0,
+      "eval_runtime": 5.7858,
+      "eval_samples_per_second": 34.049,
+      "eval_steps_per_second": 2.247,
       "step": 40
     },
     {
-      "entropy": 2.4164060140267396,
+      "entropy": 2.2022441717294545,
       "epoch": 1.202020202020202,
-      "grad_norm": 13.107892990112305,
+      "grad_norm": 2.590899705886841,
       "learning_rate": 4.1e-06,
-      "loss": 0.2526,
-      "mean_token_accuracy": 0.8814102564102564,
-      "num_tokens": 692884.0,
+      "loss": 0.0656,
+      "mean_token_accuracy": 0.9664224661313571,
+      "num_tokens": 702354.0,
       "step": 60
     },
     {
       "epoch": 1.202020202020202,
-      "eval_entropy": 2.394979641987727,
-      "eval_loss": 0.2977635860443115,
-      "eval_mean_token_accuracy": 0.8701923076923077,
-      "eval_num_tokens": 692884.0,
-      "eval_runtime": 5.7325,
-      "eval_samples_per_second": 34.366,
-      "eval_steps_per_second": 2.268,
+      "eval_entropy": 2.1788861201359677,
+      "eval_loss": 0.08057427406311035,
+      "eval_mean_token_accuracy": 0.9635989024088933,
+      "eval_num_tokens": 702354.0,
+      "eval_runtime": 5.7897,
+      "eval_samples_per_second": 34.026,
+      "eval_steps_per_second": 2.245,
       "step": 60
     },
     {
-      "entropy": 2.4086058020591734,
+      "entropy": 2.19284747838974,
       "epoch": 1.606060606060606,
-      "grad_norm": 5.737631797790527,
+      "grad_norm": 2.1625924110412598,
       "learning_rate": 2.1000000000000002e-06,
-      "loss": 0.2291,
-      "mean_token_accuracy": 0.8953125,
-      "num_tokens": 924984.0,
+      "loss": 0.0579,
+      "mean_token_accuracy": 0.9754464238882065,
+      "num_tokens": 937654.0,
       "step": 80
     },
     {
       "epoch": 1.606060606060606,
-      "eval_entropy": 2.3859705924987793,
-      "eval_loss": 0.2634693384170532,
-      "eval_mean_token_accuracy": 0.8725961538461539,
-      "eval_num_tokens": 924984.0,
-      "eval_runtime": 5.7372,
-      "eval_samples_per_second": 34.337,
-      "eval_steps_per_second": 2.266,
+      "eval_entropy": 2.178425770539504,
+      "eval_loss": 0.07602041214704514,
+      "eval_mean_token_accuracy": 0.965659343279325,
+      "eval_num_tokens": 937654.0,
+      "eval_runtime": 5.786,
+      "eval_samples_per_second": 34.047,
+      "eval_steps_per_second": 2.247,
       "step": 80
+    },
+    {
+      "entropy": 2.1867658969683523,
+      "epoch": 2.0,
+      "grad_norm": 1.2462379932403564,
+      "learning_rate": 1.0000000000000001e-07,
+      "loss": 0.0512,
+      "mean_token_accuracy": 0.9764194106444334,
+      "num_tokens": 1165538.0,
+      "step": 100
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 2.181581442172711,
+      "eval_loss": 0.0743941143155098,
+      "eval_mean_token_accuracy": 0.963598906993866,
+      "eval_num_tokens": 1165538.0,
+      "eval_runtime": 5.7817,
+      "eval_samples_per_second": 34.073,
+      "eval_steps_per_second": 2.248,
+      "step": 100
     }
   ],
   "logging_steps": 20,
@@ -106,12 +127,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 1.5898111053751296e+16,
+  "total_flos": 1.9940993720561664e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c58200221e48a21d7c48e6e07381e8bf36c583d36e4d40fe8b11bfda35fca12
+oid sha256:7330fa1bb4c98d5e0bead323a45e059f49e51c7dd2d39c91a75a154d4bfb59b4
 size 6289