Model save

README.md

@@ -4,7 +4,6 @@ library_name: transformers
 model_name: llama_check_tuned
 tags:
 - generated_from_trainer
-- alignment-handbook
 - trl
 - sft
 licence: license

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 2.0,
+    "epoch": 8.0,
     "eval_loss": 1.131566047668457,
     "eval_runtime": 0.1689,
     "eval_samples": 10,
     "eval_samples_per_second": 59.209,
     "eval_steps_per_second": 11.842,
-    "total_flos": 239345474732032.0,
-    "train_loss": 1.1170939207077026,
-    "train_runtime": 1.5876,
+    "total_flos": 981465273729024.0,
+    "train_loss": 1.1090886294841766,
+    "train_runtime": 6.846,
     "train_samples": 10,
-    "train_samples_per_second": 12.597,
-    "train_steps_per_second": 1.26
+    "train_samples_per_second": 11.686,
+    "train_steps_per_second": 1.169
 }

config.json

@@ -36,6 +36,6 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.46.3",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 128256
 }

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13cbd6d16e927a0c5bad54102514e6e18b4a47b3a6eb911e39d678d328d19f55
+oid sha256:8f407073dc27d61cb5af438c45bbd0344ad11c8a240c4d87938d053193ed3846
 size 4965799096

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f66ee63b3de299eb9fd89dfbc23a39095197874fd6891f7d18de34c50e4e1a7
+oid sha256:7e29f2c22d9ad84bbf25fafb8f5a132bbdab172fbfd9be8639621b96a0a470a0
 size 2247734992

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 2.0,
-    "total_flos": 239345474732032.0,
-    "train_loss": 1.1170939207077026,
-    "train_runtime": 1.5876,
+    "epoch": 8.0,
+    "total_flos": 981465273729024.0,
+    "train_loss": 1.1090886294841766,
+    "train_runtime": 6.846,
     "train_samples": 10,
-    "train_samples_per_second": 12.597,
-    "train_steps_per_second": 1.26
+    "train_samples_per_second": 11.686,
+    "train_steps_per_second": 1.169
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
+  "epoch": 8.0,
   "eval_steps": 500,
-  "global_step": 2,
+  "global_step": 8,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -17,39 +17,128 @@
     {
       "epoch": 1.0,
       "eval_loss": 1.131566047668457,
-      "eval_runtime": 0.1556,
-      "eval_samples_per_second": 64.262,
-      "eval_steps_per_second": 12.852,
+      "eval_runtime": 0.1572,
+      "eval_samples_per_second": 63.617,
+      "eval_steps_per_second": 12.723,
       "step": 1
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.0,
+      "learning_rate": 1.9009688679024189e-06,
       "loss": 1.1065,
       "step": 2
     },
     {
       "epoch": 2.0,
       "eval_loss": 1.131566047668457,
-      "eval_runtime": 0.1559,
-      "eval_samples_per_second": 64.123,
-      "eval_steps_per_second": 12.825,
+      "eval_runtime": 0.155,
+      "eval_samples_per_second": 64.522,
+      "eval_steps_per_second": 12.904,
       "step": 2
     },
     {
-      "epoch": 2.0,
-      "step": 2,
-      "total_flos": 239345474732032.0,
-      "train_loss": 1.1170939207077026,
-      "train_runtime": 1.5876,
-      "train_samples_per_second": 12.597,
-      "train_steps_per_second": 1.26
+      "epoch": 3.0,
+      "learning_rate": 1.6234898018587336e-06,
+      "loss": 1.1262,
+      "step": 3
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.131566047668457,
+      "eval_runtime": 0.1549,
+      "eval_samples_per_second": 64.547,
+      "eval_steps_per_second": 12.909,
+      "step": 3
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 9.71204948425293,
+      "learning_rate": 1.2225209339563143e-06,
+      "loss": 1.122,
+      "step": 4
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.1228824853897095,
+      "eval_runtime": 0.1564,
+      "eval_samples_per_second": 63.951,
+      "eval_steps_per_second": 12.79,
+      "step": 4
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 9.71204948425293,
+      "learning_rate": 7.774790660436857e-07,
+      "loss": 1.1074,
+      "step": 5
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.1228824853897095,
+      "eval_runtime": 0.1547,
+      "eval_samples_per_second": 64.658,
+      "eval_steps_per_second": 12.932,
+      "step": 5
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 9.71204948425293,
+      "learning_rate": 3.765101981412665e-07,
+      "loss": 1.0914,
+      "step": 6
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.1228824853897095,
+      "eval_runtime": 0.1548,
+      "eval_samples_per_second": 64.58,
+      "eval_steps_per_second": 12.916,
+      "step": 6
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 9.71204948425293,
+      "learning_rate": 9.903113209758096e-08,
+      "loss": 1.0813,
+      "step": 7
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.1228824853897095,
+      "eval_runtime": 0.1545,
+      "eval_samples_per_second": 64.709,
+      "eval_steps_per_second": 12.942,
+      "step": 7
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 9.079455375671387,
+      "learning_rate": 0.0,
+      "loss": 1.1103,
+      "step": 8
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 1.1230531930923462,
+      "eval_runtime": 0.1545,
+      "eval_samples_per_second": 64.744,
+      "eval_steps_per_second": 12.949,
+      "step": 8
+    },
+    {
+      "epoch": 8.0,
+      "step": 8,
+      "total_flos": 981465273729024.0,
+      "train_loss": 1.1090886294841766,
+      "train_runtime": 6.846,
+      "train_samples_per_second": 11.686,
+      "train_steps_per_second": 1.169
     }
   ],
   "logging_steps": 1,
-  "max_steps": 2,
+  "max_steps": 8,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -63,7 +152,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 239345474732032.0,
+  "total_flos": 981465273729024.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bfdeb31251f018eed190305404c961d1de16f1d879606964b08caa9ae1fcd9a
+oid sha256:033e0f2dc57707a8efb6a46923345fabb007786fea52c40f797b227fcda45a7a
 size 6968