End of training

README.md

@@ -2,13 +2,13 @@
 license: gemma
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - sft
-- alignment-handbook
 - 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: []
@@ -19,9 +19,9 @@ 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: 2.7154
+- Loss: 2.7177
 
 ## Model description
 

all_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 13.333333333333334,
-    "eval_loss": 2.71852707862854,
-    "eval_runtime": 0.4953,
+    "eval_loss": 2.7176907062530518,
+    "eval_runtime": 0.8578,
     "eval_samples": 25,
-    "eval_samples_per_second": 42.399,
-    "eval_steps_per_second": 2.019,
+    "eval_samples_per_second": 24.482,
+    "eval_steps_per_second": 1.166,
     "total_flos": 1.1018283868225536e+18,
     "train_loss": 0.0,
     "train_runtime": 3.5395,

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 14.857142857142858,
-    "eval_loss": 2.71852707862854,
-    "eval_runtime": 0.4953,
+    "epoch": 13.333333333333334,
+    "eval_loss": 2.7176907062530518,
+    "eval_runtime": 0.8578,
     "eval_samples": 25,
-    "eval_samples_per_second": 42.399,
-    "eval_steps_per_second": 2.019
+    "eval_samples_per_second": 24.482,
+    "eval_steps_per_second": 1.166
 }

runs/Jun10_04-56-56_48ddfe8e991f/events.out.tfevents.1717995445.48ddfe8e991f.131950.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e2ece50738e5f6f98d84a827912d2ed7a57a14937b046cf1417d3039b9bc4a6
+size 359