Checkpoint after epoch 1

README.md

@@ -4,6 +4,8 @@ license: llama3.2
 base_model: meta-llama/Llama-3.2-1B
 tags:
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: test
   results: []
@@ -15,6 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 # test
 
 This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.6568
+- Accuracy: 0.5231
 
 ## Model description
 

all_results.json

@@ -1,16 +1,14 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.5230333635625984,
-    "eval_loss": 2.656813859939575,
-    "eval_runtime": 59.9412,
-    "eval_samples": 5125,
-    "eval_samples_per_second": 85.5,
-    "eval_steps_per_second": 10.694,
-    "perplexity": 14.250811606566634,
+    "eval_accuracy": 0.5230967495584936,
+    "eval_loss": 2.656780958175659,
+    "eval_runtime": 59.9656,
+    "eval_samples_per_second": 85.466,
+    "eval_steps_per_second": 10.689,
+    "perplexity": 14.250342737440896,
     "total_flos": 1695049253388288.0,
-    "train_loss": 2.6765631017550615,
-    "train_runtime": 106.623,
-    "train_samples": 567,
-    "train_samples_per_second": 5.318,
-    "train_steps_per_second": 1.332
+    "train_loss": 2.676516035912742,
+    "train_runtime": 105.7579,
+    "train_samples_per_second": 5.361,
+    "train_steps_per_second": 1.343
 }

eval_epoch_1_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.5230333635625984,
-    "eval_loss": 2.656813859939575,
-    "eval_runtime": 59.8639,
-    "eval_samples_per_second": 85.611,
-    "eval_steps_per_second": 10.708,
-    "perplexity": 14.250811606566634
+    "eval_accuracy": 0.5230967495584936,
+    "eval_loss": 2.656780958175659,
+    "eval_runtime": 59.9656,
+    "eval_samples_per_second": 85.466,
+    "eval_steps_per_second": 10.689,
+    "perplexity": 14.250342737440896
 }

special_tokens_map.json

@@ -12,5 +12,6 @@
     "normalized": false,
     "rstrip": false,
     "single_word": false
-  }
+  },
+  "pad_token": "<|end_of_text|>"
 }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
-size 17209920
+oid sha256:8455cd05329f9cb9895e200605934b714e7fc55873b9969c77d1c3c01ccc60e0
+size 17210188

tokenizer_config.json

@@ -2058,5 +2058,6 @@
     "attention_mask"
   ],
   "model_max_length": 131072,
+  "pad_token": "<|end_of_text|>",
   "tokenizer_class": "PreTrainedTokenizerFast"
 }

train_epoch_1_results.json

@@ -1,9 +1,8 @@
 {
     "epoch": 1.0,
     "total_flos": 1695049253388288.0,
-    "train_loss": 2.6765631017550615,
-    "train_runtime": 106.623,
-    "train_samples": 567,
-    "train_samples_per_second": 5.318,
-    "train_steps_per_second": 1.332
+    "train_loss": 2.676516035912742,
+    "train_runtime": 105.7579,
+    "train_samples_per_second": 5.361,
+    "train_steps_per_second": 1.343
 }

trainer_state.json

@@ -11,30 +11,21 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5230333635625984,
-      "eval_loss": 2.656813859939575,
-      "eval_runtime": 59.8708,
-      "eval_samples_per_second": 85.601,
-      "eval_steps_per_second": 10.706,
+      "eval_accuracy": 0.5230967495584936,
+      "eval_loss": 2.656780958175659,
+      "eval_runtime": 59.6737,
+      "eval_samples_per_second": 85.884,
+      "eval_steps_per_second": 10.742,
       "step": 142
     },
     {
       "epoch": 1.0,
       "step": 142,
       "total_flos": 1695049253388288.0,
-      "train_loss": 2.6765631017550615,
-      "train_runtime": 106.623,
-      "train_samples_per_second": 5.318,
-      "train_steps_per_second": 1.332
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.5230333635625984,
-      "eval_loss": 2.656813859939575,
-      "eval_runtime": 59.8639,
-      "eval_samples_per_second": 85.611,
-      "eval_steps_per_second": 10.708,
-      "step": 142
+      "train_loss": 2.676516035912742,
+      "train_runtime": 105.7579,
+      "train_samples_per_second": 5.361,
+      "train_steps_per_second": 1.343
     }
   ],
   "logging_steps": 500,