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: 2.5322
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 9.904761904761905,
+    "eval_loss": 2.532215118408203,
+    "eval_runtime": 0.49,
+    "eval_samples": 25,
+    "eval_samples_per_second": 20.409,
+    "eval_steps_per_second": 2.041,
     "total_flos": 8.265211656850637e+17,
     "train_loss": 1.0387523100926326,
     "train_runtime": 3563.5293,

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.904761904761905,
-    "eval_loss": 2.5155773162841797,
-    "eval_runtime": 0.5004,
+    "eval_loss": 2.532215118408203,
+    "eval_runtime": 0.49,
     "eval_samples": 25,
-    "eval_samples_per_second": 19.982,
-    "eval_steps_per_second": 1.998
+    "eval_samples_per_second": 20.409,
+    "eval_steps_per_second": 2.041
 }

runs/Jun10_06-37-27_48ddfe8e991f/events.out.tfevents.1718005048.48ddfe8e991f.199825.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:407e1353978c1afb35d6f9a44693951439c1bba0099998b20a72d6145004a989
+size 359