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-2b
 datasets:
-- generator
+- llama-duo/synth_summarize_dataset_dedup
 model-index:
 - name: gemma2b-summarize-gemini1_5flash-64k
   results: []
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma2b-summarize-gemini1_5flash-64k
 
-This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
+This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the llama-duo/synth_summarize_dataset_dedup dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.5156
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 9.904761904761905,
+    "eval_loss": 2.5155773162841797,
+    "eval_runtime": 0.5004,
+    "eval_samples": 25,
+    "eval_samples_per_second": 19.982,
+    "eval_steps_per_second": 1.998,
     "total_flos": 8.18501087353897e+17,
     "train_loss": 1.098491345002101,
     "train_runtime": 2780.3493,

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 13.333333333333334,
-    "eval_loss": 2.7176907062530518,
-    "eval_runtime": 0.8578,
+    "epoch": 9.904761904761905,
+    "eval_loss": 2.5155773162841797,
+    "eval_runtime": 0.5004,
     "eval_samples": 25,
-    "eval_samples_per_second": 24.482,
-    "eval_steps_per_second": 1.166
+    "eval_samples_per_second": 19.982,
+    "eval_steps_per_second": 1.998
 }

runs/Jun10_05-12-11_48ddfe8e991f/events.out.tfevents.1717999157.48ddfe8e991f.142677.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0a679e5660758c302cd622f9cddc78b180a47bf55b158baab54ae8cf2d1a0c6
+size 359