End of training

README.md

@@ -2,12 +2,13 @@
 license: gemma
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
 base_model: google/gemma-7b
 datasets:
-- generator
+- llama-duo/synth_summarize_dataset
 model-index:
 - name: gemma7b-summarize-claude3sonnet-30k
   results: []
@@ -19,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/chansung18/huggingface/runs/7bdtvabz)
 # gemma7b-summarize-claude3sonnet-30k
 
-This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
+This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the llama-duo/synth_summarize_dataset dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.0044
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 10.0,
+    "eval_loss": 3.0043678283691406,
+    "eval_runtime": 1.0435,
+    "eval_samples": 25,
+    "eval_samples_per_second": 4.792,
+    "eval_steps_per_second": 1.917,
     "total_flos": 2.2627322959180595e+18,
     "train_loss": 1.3636517344294368,
     "train_runtime": 11668.1199,

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 5.0,
-    "eval_loss": 2.326845407485962,
-    "eval_runtime": 1.0399,
+    "epoch": 10.0,
+    "eval_loss": 3.0043678283691406,
+    "eval_runtime": 1.0435,
     "eval_samples": 25,
-    "eval_samples_per_second": 4.808,
-    "eval_steps_per_second": 1.923
+    "eval_samples_per_second": 4.792,
+    "eval_steps_per_second": 1.917
 }

runs/May21_20-36-36_deep-diver-main-dangerous-sloth-1-0-0/events.out.tfevents.1716350015.deep-diver-main-dangerous-sloth-1-0-0.385.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:544b831c85a8b924dabc7295519fe4a9adbd28c750534f60e8ceda2031d35e3c
+size 359