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: []
@@ -18,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](None)
 # 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: 1.0774
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 2.994991652754591,
+    "eval_loss": 1.0773704051971436,
+    "eval_runtime": 21.3034,
+    "eval_samples": 500,
+    "eval_samples_per_second": 89.798,
+    "eval_steps_per_second": 2.816,
     "total_flos": 246978202042368.0,
     "train_loss": 0.8908544659880359,
     "train_runtime": 5202.4388,

config.json

@@ -24,6 +24,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.42.3",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.994991652754591,
+    "eval_loss": 1.0773704051971436,
+    "eval_runtime": 21.3034,
+    "eval_samples": 500,
+    "eval_samples_per_second": 89.798,
+    "eval_steps_per_second": 2.816
+}

runs/Jul10_15-20-59_ale-distillm-8-0-0/events.out.tfevents.1720626952.ale-distillm-8-0-0.2231.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cced5b5c4a67cf4d99305eb2b399025e98514be35116631b12f1407ac4cd3400
+size 359