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. -->
 
 # llama3-8b-classification-gpt4o-100k2
 
-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: 1.7732
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9977827050997783,
+    "eval_loss": 1.7732421159744263,
+    "eval_runtime": 0.5704,
+    "eval_samples": 16,
+    "eval_samples_per_second": 1.753,
+    "eval_steps_per_second": 1.753,
     "total_flos": 3.3259687694984806e+17,
     "train_loss": 1.4963340536753336,
     "train_runtime": 725.2803,

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.988913525498892,
-    "eval_loss": 1.9855170249938965,
-    "eval_runtime": 0.3687,
+    "epoch": 0.9977827050997783,
+    "eval_loss": 1.7732421159744263,
+    "eval_runtime": 0.5704,
     "eval_samples": 16,
-    "eval_samples_per_second": 2.712,
-    "eval_steps_per_second": 2.712
+    "eval_samples_per_second": 1.753,
+    "eval_steps_per_second": 1.753
 }

runs/Aug08_10-30-50_main-soft-horse-1-0-0/events.out.tfevents.1723128439.main-soft-horse-1-0-0.543.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:221c78480a1531d48319ffbe8c5e9e1049a4aa5fdec207446d170dbe5d47bbe6
+size 359