End of training

README.md

@@ -2,10 +2,13 @@
 license: apache-2.0
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
 base_model: mistralai/Mistral-7B-v0.1
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-7b
   results: []
@@ -16,19 +19,19 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-qlora](https://huggingface.co/alignment-handbook/zephyr-7b-sft-qlora) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6918
-- Rewards/chosen: -0.0863
-- Rewards/rejected: -0.1983
-- Rewards/accuracies: 0.3571
-- Rewards/margins: 0.1120
-- Logps/rejected: -95.2291
-- Logps/chosen: -77.5275
-- Logits/rejected: -1.9113
-- Logits/chosen: -1.9391
-- Use Label: 14335.7139
-- Pred Label: 4352.2856
+- Rewards/chosen: -0.0862
+- Rewards/rejected: -0.1980
+- Rewards/accuracies: 0.3591
+- Rewards/margins: 0.1117
+- Logps/rejected: -95.1937
+- Logps/chosen: -77.5232
+- Logits/rejected: -1.9123
+- Logits/chosen: -1.9402
+- Use Label: 15333.4131
+- Pred Label: 4738.5874
 
 ## Model description
 

all_results.json

@@ -1,5 +1,20 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -1.9401931762695312,
+    "eval_logits/rejected": -1.9123154878616333,
+    "eval_logps/chosen": -77.5232162475586,
+    "eval_logps/rejected": -95.19373321533203,
+    "eval_loss": 0.6917868852615356,
+    "eval_pred_label": 4738.58740234375,
+    "eval_rewards/accuracies": 0.3591269850730896,
+    "eval_rewards/chosen": -0.0862266793847084,
+    "eval_rewards/margins": 0.11172995716333389,
+    "eval_rewards/rejected": -0.19795666635036469,
+    "eval_runtime": 247.3331,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 8.086,
+    "eval_steps_per_second": 0.255,
+    "eval_use_label": 15333.4130859375,
     "train_loss": 0.6906769273168754,
     "train_runtime": 20027.4031,
     "train_samples": 61135,

eval_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -0.24525223672389984,
-    "eval_logits/rejected": -0.1380803883075714,
-    "eval_logps/chosen": -103.02954864501953,
-    "eval_logps/rejected": -131.91891479492188,
-    "eval_loss": 0.6906174421310425,
-    "eval_pred_label": 4192.14306640625,
-    "eval_rewards/accuracies": 0.363095223903656,
-    "eval_rewards/chosen": -0.3412899374961853,
-    "eval_rewards/margins": 0.22391849756240845,
-    "eval_rewards/rejected": -0.5652084350585938,
-    "eval_runtime": 247.5585,
+    "eval_logits/chosen": -1.9401931762695312,
+    "eval_logits/rejected": -1.9123154878616333,
+    "eval_logps/chosen": -77.5232162475586,
+    "eval_logps/rejected": -95.19373321533203,
+    "eval_loss": 0.6917868852615356,
+    "eval_pred_label": 4738.58740234375,
+    "eval_rewards/accuracies": 0.3591269850730896,
+    "eval_rewards/chosen": -0.0862266793847084,
+    "eval_rewards/margins": 0.11172995716333389,
+    "eval_rewards/rejected": -0.19795666635036469,
+    "eval_runtime": 247.3331,
     "eval_samples": 2000,
-    "eval_samples_per_second": 8.079,
-    "eval_steps_per_second": 0.254,
-    "eval_use_label": 15879.857421875
+    "eval_samples_per_second": 8.086,
+    "eval_steps_per_second": 0.255,
+    "eval_use_label": 15333.4130859375
 }