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-claude3sonnet-32k
   results: []
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma2b-summarize-claude3sonnet-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: 2.5627
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 9.950738916256158,
+    "eval_loss": 2.562716007232666,
+    "eval_runtime": 0.5025,
+    "eval_samples": 25,
+    "eval_samples_per_second": 19.901,
+    "eval_steps_per_second": 1.99,
     "total_flos": 5.9616784580975e+17,
     "train_loss": 1.0187734235631358,
     "train_runtime": 5372.8296,

config.json

@@ -9,7 +9,7 @@
   "eos_token_id": 1,
   "head_dim": 256,
   "hidden_act": "gelu",
-  "hidden_activation": null,
+  "hidden_activation": "gelu_pytorch_tanh",
   "hidden_size": 2048,
   "initializer_range": 0.02,
   "intermediate_size": 16384,
@@ -38,7 +38,7 @@
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.40.0",
+  "transformers_version": "4.41.2",
   "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.950738916256158,
-    "eval_loss": 2.523752212524414,
-    "eval_runtime": 0.4919,
+    "eval_loss": 2.562716007232666,
+    "eval_runtime": 0.5025,
     "eval_samples": 25,
-    "eval_samples_per_second": 20.328,
-    "eval_steps_per_second": 2.033
+    "eval_samples_per_second": 19.901,
+    "eval_steps_per_second": 1.99
 }

runs/Jun10_10-58-48_user-HP-Z8-Fury-G5-Workstation-Desktop-PC/events.out.tfevents.1717990137.user-HP-Z8-Fury-G5-Workstation-Desktop-PC.2790.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c15696d623e37dbd52d1eff103f2cf712e80b3eca25f90ed254a9ec5c8273fc
+size 359