End of training

README.md

@@ -3,9 +3,9 @@ base_model: unsloth/llama-3-8b-Instruct-bnb-4bit
 library_name: peft
 license: llama3
 tags:
-- unsloth
 - llama-factory
 - lora
+- unsloth
 - generated_from_trainer
 model-index:
 - name: revised-llamafactory-llama-3-8b
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # revised-llamafactory-llama-3-8b
 
-This model is a fine-tuned version of [unsloth/llama-3-8b-Instruct-bnb-4bit](https://huggingface.co/unsloth/llama-3-8b-Instruct-bnb-4bit) on an unknown dataset.
+This model is a fine-tuned version of [unsloth/llama-3-8b-Instruct-bnb-4bit](https://huggingface.co/unsloth/llama-3-8b-Instruct-bnb-4bit) on the formatted_data dataset.
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.992,
-    "total_flos": 1.7141703652933632e+16,
-    "train_loss": 0.15340985093385942,
-    "train_runtime": 956.257,
-    "train_samples_per_second": 0.523,
-    "train_steps_per_second": 0.065
+    "total_flos": 1.8193643319066624e+16,
+    "train_loss": 0.15735646361304867,
+    "train_runtime": 1055.5718,
+    "train_samples_per_second": 0.474,
+    "train_steps_per_second": 0.059
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.992,
-    "total_flos": 1.7141703652933632e+16,
-    "train_loss": 0.15340985093385942,
-    "train_runtime": 956.257,
-    "train_samples_per_second": 0.523,
-    "train_steps_per_second": 0.065
+    "total_flos": 1.8193643319066624e+16,
+    "train_loss": 0.15735646361304867,
+    "train_runtime": 1055.5718,
+    "train_samples_per_second": 0.474,
+    "train_steps_per_second": 0.059
 }

trainer_state.json

@@ -10,54 +10,54 @@
   "log_history": [
     {
       "epoch": 0.16,
-      "grad_norm": 1.2404758930206299,
+      "grad_norm": 1.192562460899353,
       "learning_rate": 1.9853538358476933e-05,
-      "loss": 0.2622,
+      "loss": 0.2839,
       "step": 10
     },
     {
       "epoch": 0.32,
-      "grad_norm": 2.941343069076538,
+      "grad_norm": 3.5562169551849365,
       "learning_rate": 1.736741137876405e-05,
-      "loss": 0.2174,
+      "loss": 0.2157,
       "step": 20
     },
     {
       "epoch": 0.48,
-      "grad_norm": 2.0719258785247803,
+      "grad_norm": 1.9338874816894531,
       "learning_rate": 1.2542183341934873e-05,
-      "loss": 0.1544,
+      "loss": 0.1568,
       "step": 30
     },
     {
       "epoch": 0.64,
-      "grad_norm": 1.2970093488693237,
+      "grad_norm": 1.4656245708465576,
       "learning_rate": 6.909830056250527e-06,
-      "loss": 0.0843,
+      "loss": 0.0839,
       "step": 40
     },
     {
       "epoch": 0.8,
-      "grad_norm": 2.1786603927612305,
+      "grad_norm": 1.7528181076049805,
       "learning_rate": 2.2585838936091753e-06,
-      "loss": 0.1001,
+      "loss": 0.1057,
       "step": 50
     },
     {
       "epoch": 0.96,
-      "grad_norm": 1.0291296243667603,
+      "grad_norm": 2.33693528175354,
       "learning_rate": 6.51826465144978e-08,
-      "loss": 0.126,
+      "loss": 0.1226,
       "step": 60
     },
     {
       "epoch": 0.992,
       "step": 62,
-      "total_flos": 1.7141703652933632e+16,
-      "train_loss": 0.15340985093385942,
-      "train_runtime": 956.257,
-      "train_samples_per_second": 0.523,
-      "train_steps_per_second": 0.065
+      "total_flos": 1.8193643319066624e+16,
+      "train_loss": 0.15735646361304867,
+      "train_runtime": 1055.5718,
+      "train_samples_per_second": 0.474,
+      "train_steps_per_second": 0.059
     }
   ],
   "logging_steps": 10,
@@ -77,7 +77,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.7141703652933632e+16,
+  "total_flos": 1.8193643319066624e+16,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null