End of training

README.md

@@ -1,17 +1,19 @@
 ---
 base_model: mistralai/Mistral-7B-Instruct-v0.2
+datasets: sum-pref-overfit
 library_name: transformers
-model_name: prometheus-7b-v1.5-beta-1-0109-sum-pref-overfit-v3-flash
+model_name: sum-pref-overfit-v1-modelcard
 tags:
 - generated_from_trainer
+- alignment-handbook
 - trl
 - sft
 licence: license
 ---
 
-# Model Card for prometheus-7b-v1.5-beta-1-0109-sum-pref-overfit-v3-flash
+# Model Card for sum-pref-overfit-v1-modelcard
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2).
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the [sum-pref-overfit](https://huggingface.co/datasets/sum-pref-overfit) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 0.010453196242451668,
+    "eval_runtime": 1.279,
+    "eval_samples": 10,
+    "eval_samples_per_second": 0.782,
+    "eval_steps_per_second": 0.782,
     "total_flos": 2.542777053216768e+16,
     "train_loss": 0.06061042585337521,
     "train_runtime": 596.3446,

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_loss": 0.01081210095435381,
-    "eval_runtime": 1.2807,
+    "eval_loss": 0.010453196242451668,
+    "eval_runtime": 1.279,
     "eval_samples": 10,
-    "eval_samples_per_second": 0.781,
-    "eval_steps_per_second": 0.781
+    "eval_samples_per_second": 0.782,
+    "eval_steps_per_second": 0.782
 }