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-gpt4o-32k
   results: []
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma2b-summarize-gpt4o-32k
 
-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: 2.6191
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 10.0,
+    "eval_loss": 2.6190853118896484,
+    "eval_runtime": 0.5288,
+    "eval_samples": 25,
+    "eval_samples_per_second": 18.91,
+    "eval_steps_per_second": 1.891,
     "total_flos": 4.3089359154446336e+17,
     "train_loss": 1.133717959547696,
     "train_runtime": 3901.5754,

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_loss": 2.5617620944976807,
-    "eval_runtime": 0.5264,
+    "eval_loss": 2.6190853118896484,
+    "eval_runtime": 0.5288,
     "eval_samples": 25,
-    "eval_samples_per_second": 18.996,
-    "eval_steps_per_second": 1.9
+    "eval_samples_per_second": 18.91,
+    "eval_steps_per_second": 1.891
 }

runs/Jun10_10-37-10_user-HP-Z8-Fury-G5-Workstation-Desktop-PC/events.out.tfevents.1717987356.user-HP-Z8-Fury-G5-Workstation-Desktop-PC.72146.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:458d1c934bfcabe6eae3a28172b18fe1c5a722671bd85e187f5e4e66797624af
+size 359