End of training

README.md

@@ -3,10 +3,16 @@ 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
 - alignment-handbook
 - generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-7b-dpo-full
   results: []
@@ -17,17 +23,17 @@ 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 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:
-- Logits/chosen: -0.3096
-- Logits/rejected: 0.6049
-- Logps/chosen: -755.9323
-- Logps/rejected: -1192.5621
-- Loss: 0.3152
-- Rewards/accuracies: 0.8184
-- Rewards/chosen: -4.6496
-- Rewards/margins: 4.3751
-- Rewards/rejected: -9.0247
+- Loss: 0.3148
+- Rewards/chosen: -4.6400
+- Rewards/rejected: -8.9971
+- Rewards/accuracies: 0.8178
+- Rewards/margins: 4.3571
+- Logps/rejected: -1189.8031
+- Logps/chosen: -754.9667
+- Logits/rejected: 0.6135
+- Logits/chosen: -0.2969
 
 ## Model description
 

all_results.json

@@ -9,10 +9,10 @@
     "eval_rewards/chosen": -4.639986038208008,
     "eval_rewards/margins": 4.357123374938965,
     "eval_rewards/rejected": -8.997109413146973,
-    "eval_runtime": 648.8061,
+    "eval_runtime": 645.0187,
     "eval_samples": 7126,
-    "eval_samples_per_second": 10.983,
-    "eval_steps_per_second": 0.344,
+    "eval_samples_per_second": 11.048,
+    "eval_steps_per_second": 0.346,
     "total_flos": 0.0,
     "train_loss": 0.0,
     "train_runtime": 0.0165,

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

@@ -9,8 +9,8 @@
     "eval_rewards/chosen": -4.639986038208008,
     "eval_rewards/margins": 4.357123374938965,
     "eval_rewards/rejected": -8.997109413146973,
-    "eval_runtime": 648.8061,
+    "eval_runtime": 645.0187,
     "eval_samples": 7126,
-    "eval_samples_per_second": 10.983,
-    "eval_steps_per_second": 0.344
+    "eval_samples_per_second": 11.048,
+    "eval_steps_per_second": 0.346
 }

runs/Aug25_00-33-37_ip-10-0-9-154.ec2.internal/events.out.tfevents.1724548152.ip-10-0-9-154.ec2.internal.32573.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b7676a18a5d12857ee15d41602aece37dabdbe0754444e4a6fd38b30ce440df
+size 828