End of training

README.md

@@ -3,11 +3,19 @@ library_name: transformers
 license: gemma
 base_model: google/gemma-2-9b
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- oxford-llms/ultrachat_filtered
+- oxford-llms/Magpie-Qwen2.5-Pro-1M-v0.1-filtered
+- oxford-llms/european_social_survey_2020_sft
+- oxford-llms/european_social_survey_2023_sft
+- oxford-llms/world_values_survey_2017_2022_sft
 model-index:
 - name: ox-llms-9b-gemma-sft
   results: []
@@ -18,7 +26,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ox-llms-9b-gemma-sft
 
-This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on the generator dataset.
+This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on the oxford-llms/ultrachat_filtered, the oxford-llms/Magpie-Qwen2.5-Pro-1M-v0.1-filtered, the oxford-llms/european_social_survey_2020_sft, the oxford-llms/european_social_survey_2023_sft and the oxford-llms/world_values_survey_2017_2022_sft datasets.
 It achieves the following results on the evaluation set:
 - Loss: 0.8087
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 2.0,
+    "eval_loss": 0.8086519241333008,
+    "eval_runtime": 263.8826,
+    "eval_samples": 41639,
+    "eval_samples_per_second": 31.313,
+    "eval_steps_per_second": 0.981,
     "total_flos": 1845812527104000.0,
     "train_loss": 0.6640872944247338,
     "train_runtime": 43998.9378,

config.json

@@ -29,6 +29,6 @@
   "sliding_window_size": 4096,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.45.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 0.8086519241333008,
+    "eval_runtime": 263.8826,
+    "eval_samples": 41639,
+    "eval_samples_per_second": 31.313,
+    "eval_steps_per_second": 0.981
+}