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-align-scan-0.0-0.0-polynomial-2
   results: []
@@ -17,17 +23,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-align-scan-0.0-0.0-polynomial-2
 
-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:
-- Logits/chosen: -2.4166
-- Logits/rejected: -2.3982
-- Logps/chosen: -79.0868
-- Logps/rejected: -91.3032
-- Loss: 0.6444
-- Rewards/accuracies: 0.3234
-- Rewards/chosen: -0.1722
-- Rewards/margins: 0.2091
-- Rewards/rejected: -0.3813
+- Loss: 0.6385
+- Rewards/chosen: -0.2230
+- Rewards/rejected: -0.5393
+- Rewards/accuracies: 0.3333
+- Rewards/margins: 0.3162
+- Logps/rejected: -96.4576
+- Logps/chosen: -80.8310
+- Logits/rejected: -2.3598
+- Logits/chosen: -2.3790
 
 ## Model description
 

all_results.json

@@ -4,15 +4,15 @@
     "eval_logits/rejected": -2.3597521781921387,
     "eval_logps/chosen": -80.83098602294922,
     "eval_logps/rejected": -96.45763397216797,
-    "eval_loss": 0.6398673057556152,
+    "eval_loss": 0.6385183930397034,
     "eval_rewards/accuracies": 0.3333333432674408,
-    "eval_rewards/chosen": -0.23755605518817902,
-    "eval_rewards/margins": 0.33684492111206055,
-    "eval_rewards/rejected": -0.5744009017944336,
-    "eval_runtime": 113.9474,
+    "eval_rewards/chosen": -0.22302590310573578,
+    "eval_rewards/margins": 0.31624171137809753,
+    "eval_rewards/rejected": -0.5392675995826721,
+    "eval_runtime": 115.9806,
     "eval_samples": 2000,
-    "eval_samples_per_second": 17.552,
-    "eval_steps_per_second": 0.553,
+    "eval_samples_per_second": 17.244,
+    "eval_steps_per_second": 0.543,
     "total_flos": 0.0,
     "train_loss": 0.0,
     "train_runtime": 0.0564,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -4,13 +4,13 @@
     "eval_logits/rejected": -2.3597521781921387,
     "eval_logps/chosen": -80.83098602294922,
     "eval_logps/rejected": -96.45763397216797,
-    "eval_loss": 0.6398673057556152,
+    "eval_loss": 0.6385183930397034,
     "eval_rewards/accuracies": 0.3333333432674408,
-    "eval_rewards/chosen": -0.23755605518817902,
-    "eval_rewards/margins": 0.33684492111206055,
-    "eval_rewards/rejected": -0.5744009017944336,
-    "eval_runtime": 113.9474,
+    "eval_rewards/chosen": -0.22302590310573578,
+    "eval_rewards/margins": 0.31624171137809753,
+    "eval_rewards/rejected": -0.5392675995826721,
+    "eval_runtime": 115.9806,
     "eval_samples": 2000,
-    "eval_samples_per_second": 17.552,
-    "eval_steps_per_second": 0.553
+    "eval_samples_per_second": 17.244,
+    "eval_steps_per_second": 0.543
 }

runs/Sep28_08-14-44_qa-a100-005.crc.nd.edu/events.out.tfevents.1727526141.qa-a100-005.crc.nd.edu.126407.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfd3afd18e75cedaeb146f0ca47f265b97d9fefdb612e65bdfddb98d6b4b1644
+size 828