End of training

README.md

@@ -3,9 +3,15 @@ library_name: transformers
 license: apache-2.0
 base_model: alignment-handbook/zephyr-7b-sft-full
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-7b-group-dpo-full-4
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-group-dpo-full-4
 
-This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the None dataset.
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.3329
-- Rewards/chosen: -3.5971
-- Rewards/rejected: -6.6557
+- Rewards/chosen: -3.5718
+- Rewards/rejected: -6.6220
 - Rewards/accuracies: 0.8281
-- Rewards/margins: 3.0586
-- Logps/rejected: -955.4623
-- Logps/chosen: -642.2396
-- Logits/rejected: -0.6052
-- Logits/chosen: -1.1035
+- Rewards/margins: 3.0502
+- Logps/rejected: -952.0885
+- Logps/chosen: -639.7113
+- Logits/rejected: -0.6186
+- Logits/chosen: -1.1189
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9994242947610823,
+    "eval_logits/chosen": -1.1188652515411377,
+    "eval_logits/rejected": -0.6186092495918274,
+    "eval_logps/chosen": -639.7113037109375,
+    "eval_logps/rejected": -952.0885009765625,
+    "eval_loss": 0.3328971564769745,
+    "eval_rewards/accuracies": 0.828125,
+    "eval_rewards/chosen": -3.571798324584961,
+    "eval_rewards/margins": 3.05015230178833,
+    "eval_rewards/rejected": -6.621951103210449,
+    "eval_runtime": 330.4232,
+    "eval_samples": 7126,
+    "eval_samples_per_second": 21.566,
+    "eval_steps_per_second": 0.339,
     "total_flos": 0.0,
     "train_loss": 0.17276417467451316,
     "train_runtime": 14185.6914,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9994242947610823,
+    "eval_logits/chosen": -1.1188652515411377,
+    "eval_logits/rejected": -0.6186092495918274,
+    "eval_logps/chosen": -639.7113037109375,
+    "eval_logps/rejected": -952.0885009765625,
+    "eval_loss": 0.3328971564769745,
+    "eval_rewards/accuracies": 0.828125,
+    "eval_rewards/chosen": -3.571798324584961,
+    "eval_rewards/margins": 3.05015230178833,
+    "eval_rewards/rejected": -6.621951103210449,
+    "eval_runtime": 330.4232,
+    "eval_samples": 7126,
+    "eval_samples_per_second": 21.566,
+    "eval_steps_per_second": 0.339
+}