End of training

README.md

@@ -1,10 +1,11 @@
 ---
 base_model: meta-llama/Meta-Llama-3-8B
 datasets:
-- generator
+- llama-duo/synth_classification_dataset_dedup
 library_name: peft
 license: llama3
 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-gpt4o_100k_classification-k
 
-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 generator 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 llama-duo/synth_classification_dataset_dedup dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.6382
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9957805907172996,
+    "eval_loss": 1.6381601095199585,
+    "eval_runtime": 1.1447,
+    "eval_samples": 16,
+    "eval_samples_per_second": 1.747,
+    "eval_steps_per_second": 0.874,
     "total_flos": 3.48383355932246e+17,
     "train_loss": 1.291591884726185,
     "train_runtime": 1283.2771,

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.9957805907172996,
-    "eval_loss": 1.6322884559631348,
-    "eval_runtime": 1.1485,
+    "eval_loss": 1.6381601095199585,
+    "eval_runtime": 1.1447,
     "eval_samples": 16,
-    "eval_samples_per_second": 1.741,
-    "eval_steps_per_second": 0.871
+    "eval_samples_per_second": 1.747,
+    "eval_steps_per_second": 0.874
 }

runs/Sep29_04-53-41_82b7fdadb1f8/events.out.tfevents.1727587207.82b7fdadb1f8.364615.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4a8ce256f5774d78f4e9f1bc17f4a9199fd095038392545f8cbe53fdd8fa51c
+size 354