End of training

README.md

@@ -1,10 +1,11 @@
 ---
 base_model: meta-llama/Meta-Llama-3.1-8B
 datasets:
-- generator
+- llama-duo/synth_classification_dataset_dedup
 library_name: peft
 license: llama3.1
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3.1-8b-classification-gpt4o-100k
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the llama-duo/synth_classification_dataset_dedup dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.0330
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 10.0,
+    "eval_loss": 3.0329792499542236,
+    "eval_runtime": 0.4341,
+    "eval_samples": 16,
+    "eval_samples_per_second": 4.607,
+    "eval_steps_per_second": 2.303,
     "total_flos": 4.416382035459834e+18,
     "train_loss": 0.922980490487975,
     "train_runtime": 12382.7598,

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.977827050997783,
-    "eval_loss": 1.8520119190216064,
-    "eval_runtime": 0.3553,
+    "epoch": 10.0,
+    "eval_loss": 3.0329792499542236,
+    "eval_runtime": 0.4341,
     "eval_samples": 16,
-    "eval_samples_per_second": 2.814,
-    "eval_steps_per_second": 2.814
+    "eval_samples_per_second": 4.607,
+    "eval_steps_per_second": 2.303
 }

runs/Aug07_09-32-26_deep-diver-main-weak-ape-1-0-0/events.out.tfevents.1723050187.deep-diver-main-weak-ape-1-0-0.531.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae9bb2c7ad7d9ed7fdee930ed30ad79a06eab8961667d1b6c96f2278219ff488
+size 359