End of training

README.md

@@ -2,11 +2,15 @@
 license: gemma
 base_model: google/gemma-7b
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- HuggingFaceH4/deita-10k-v0-sft
 model-index:
 - name: zephyr-7b-gemma-sft
   results: []
@@ -17,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-gemma-sft
 
-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 HuggingFaceH4/deita-10k-v0-sft dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.0814
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 2.994991652754591,
+    "eval_loss": 1.0814331769943237,
+    "eval_runtime": 50.5833,
+    "eval_samples": 500,
+    "eval_samples_per_second": 37.819,
+    "eval_steps_per_second": 1.186,
     "total_flos": 246978202042368.0,
     "train_loss": 0.8794620578637224,
     "train_runtime": 12081.5129,

config.json

@@ -24,6 +24,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.40.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.994991652754591,
+    "eval_loss": 1.0814331769943237,
+    "eval_runtime": 50.5833,
+    "eval_samples": 500,
+    "eval_samples_per_second": 37.819,
+    "eval_steps_per_second": 1.186
+}