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-gpt4o-80k
   results: []
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma7b-summarize-gpt4o-80k
 
-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: 5.1111
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 15.0,
+    "eval_loss": 5.111132621765137,
+    "eval_runtime": 1.045,
+    "eval_samples": 25,
+    "eval_samples_per_second": 4.785,
+    "eval_steps_per_second": 1.914,
     "total_flos": 2.545573832974926e+18,
     "train_loss": 1.4823458194016694,
     "train_runtime": 13248.7618,

config.json

@@ -38,7 +38,7 @@
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.41.0",
+  "transformers_version": "4.41.1",
   "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.98185117967332,
-    "eval_loss": 2.9801077842712402,
-    "eval_runtime": 1.0531,
+    "epoch": 15.0,
+    "eval_loss": 5.111132621765137,
+    "eval_runtime": 1.045,
     "eval_samples": 25,
-    "eval_samples_per_second": 4.748,
-    "eval_steps_per_second": 1.899
+    "eval_samples_per_second": 4.785,
+    "eval_steps_per_second": 1.914
 }

runs/May23_00-47-13_deep-diver-main-splendid-ape-1-0-0/events.out.tfevents.1716453142.deep-diver-main-splendid-ape-1-0-0.385.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c14dec7ec6e62608cdd5b686c33099ebd6c34a15bf14876f92a2b670ebd5177
+size 359