End of training

README.md

@@ -2,9 +2,15 @@
 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-dpo-full
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-dpo-full
 
-This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on an unknown 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: 0.5032
-- Rewards/chosen: -1.0014
-- Rewards/rejected: -1.9606
-- Rewards/accuracies: 0.7734
-- Rewards/margins: 0.9592
-- Logps/rejected: -458.7184
-- Logps/chosen: -362.7651
-- Logits/rejected: 1.3908
-- Logits/chosen: 0.5145
+- Loss: 0.5025
+- Rewards/chosen: -1.0211
+- Rewards/rejected: -2.0056
+- Rewards/accuracies: 0.7773
+- Rewards/margins: 0.9845
+- Logps/rejected: -463.2262
+- Logps/chosen: -364.7427
+- Logits/rejected: 1.5124
+- Logits/chosen: 0.6094
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": 0.6094364523887634,
+    "eval_logits/rejected": 1.5124281644821167,
+    "eval_logps/chosen": -364.74273681640625,
+    "eval_logps/rejected": -463.2261962890625,
+    "eval_loss": 0.5025366544723511,
+    "eval_rewards/accuracies": 0.77734375,
+    "eval_rewards/chosen": -1.021127700805664,
+    "eval_rewards/margins": 0.9845134615898132,
+    "eval_rewards/rejected": -2.005641222000122,
+    "eval_runtime": 109.2669,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 18.304,
+    "eval_steps_per_second": 0.293,
     "total_flos": 0.0,
     "train_loss": 0.5406385645208,
     "train_runtime": 10904.9554,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": 0.6094364523887634,
+    "eval_logits/rejected": 1.5124281644821167,
+    "eval_logps/chosen": -364.74273681640625,
+    "eval_logps/rejected": -463.2261962890625,
+    "eval_loss": 0.5025366544723511,
+    "eval_rewards/accuracies": 0.77734375,
+    "eval_rewards/chosen": -1.021127700805664,
+    "eval_rewards/margins": 0.9845134615898132,
+    "eval_rewards/rejected": -2.005641222000122,
+    "eval_runtime": 109.2669,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 18.304,
+    "eval_steps_per_second": 0.293
+}

runs/Aug15_19-19-02_a100-st-p4de24xlarge-195/events.out.tfevents.1723765429.a100-st-p4de24xlarge-195.1070198.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0249ec813118a334f83490c2d97e2960d5e4af2809f686fefc0e745aeb7af3aa
+size 828