End of training

README.md

@@ -4,6 +4,7 @@ license: llama3
 base_model: meta-llama/Meta-Llama-3-8B
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: model
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # model
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the PrincetonPLI/Instruct-SkillMix-SDD dataset.
 It achieves the following results on the evaluation set:
 - Loss: nan
 

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 0.13043478260869565,
-    "eval_loss": 1.6954997777938843,
-    "eval_runtime": 2.0177,
-    "eval_samples_per_second": 305.789,
-    "eval_steps_per_second": 4.956,
-    "total_flos": 3.355230384370483e+16,
-    "train_loss": 1.413570721944173,
-    "train_runtime": 204.0944,
-    "train_samples_per_second": 7.526,
+    "epoch": 2.4,
+    "eval_loss": NaN,
+    "eval_runtime": 2.195,
+    "eval_samples_per_second": 46.014,
+    "eval_steps_per_second": 0.911,
+    "total_flos": 4.249521430997238e+17,
+    "train_loss": 51.85570949978299,
+    "train_runtime": 609.9474,
+    "train_samples_per_second": 9.439,
     "train_steps_per_second": 0.015
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 0.13043478260869565,
-    "eval_loss": 1.6954997777938843,
-    "eval_runtime": 2.0177,
-    "eval_samples_per_second": 305.789,
-    "eval_steps_per_second": 4.956
+    "epoch": 2.4,
+    "eval_loss": NaN,
+    "eval_runtime": 2.195,
+    "eval_samples_per_second": 46.014,
+    "eval_steps_per_second": 0.911
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 0.13043478260869565,
-    "total_flos": 3.355230384370483e+16,
-    "train_loss": 1.413570721944173,
-    "train_runtime": 204.0944,
-    "train_samples_per_second": 7.526,
+    "epoch": 2.4,
+    "total_flos": 4.249521430997238e+17,
+    "train_loss": 51.85570949978299,
+    "train_runtime": 609.9474,
+    "train_samples_per_second": 9.439,
     "train_steps_per_second": 0.015
 }

trainer_state.json

@@ -1,35 +1,51 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.13043478260869565,
+  "epoch": 2.4,
   "eval_steps": 500,
-  "global_step": 3,
+  "global_step": 9,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.13043478260869565,
-      "eval_loss": 1.6954997777938843,
-      "eval_runtime": 1.8869,
-      "eval_samples_per_second": 326.995,
-      "eval_steps_per_second": 5.3,
+      "epoch": 0.8,
+      "eval_loss": NaN,
+      "eval_runtime": 2.4966,
+      "eval_samples_per_second": 40.454,
+      "eval_steps_per_second": 0.801,
       "step": 3
     },
     {
-      "epoch": 0.13043478260869565,
-      "step": 3,
-      "total_flos": 3.355230384370483e+16,
-      "train_loss": 1.413570721944173,
-      "train_runtime": 204.0944,
-      "train_samples_per_second": 7.526,
+      "epoch": 1.8666666666666667,
+      "eval_loss": NaN,
+      "eval_runtime": 2.4813,
+      "eval_samples_per_second": 40.705,
+      "eval_steps_per_second": 0.806,
+      "step": 7
+    },
+    {
+      "epoch": 2.4,
+      "eval_loss": NaN,
+      "eval_runtime": 1.9517,
+      "eval_samples_per_second": 51.75,
+      "eval_steps_per_second": 1.025,
+      "step": 9
+    },
+    {
+      "epoch": 2.4,
+      "step": 9,
+      "total_flos": 4.249521430997238e+17,
+      "train_loss": 51.85570949978299,
+      "train_runtime": 609.9474,
+      "train_samples_per_second": 9.439,
       "train_steps_per_second": 0.015
     }
   ],
   "logging_steps": 10,
-  "max_steps": 3,
+  "max_steps": 9,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -43,8 +59,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.355230384370483e+16,
-  "train_batch_size": 16,
+  "total_flos": 4.249521430997238e+17,
+  "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null
 }