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-align-scan-7e-07-0.45-cosine-3.0
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-align-scan-7e-07-0.45-cosine-3.0
 
-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: 0.9344
-- Rewards/chosen: -0.3373
-- Rewards/rejected: -2.0235
-- Rewards/accuracies: 0.3452
-- Rewards/margins: 1.6862
-- Logps/rejected: -85.6249
-- Logps/chosen: -75.2407
+- Loss: 0.9379
+- Rewards/chosen: -0.3417
+- Rewards/rejected: -2.0404
+- Rewards/accuracies: 0.3472
+- Rewards/margins: 1.6986
+- Logps/rejected: -85.6625
+- Logps/chosen: -75.2506
 - Logits/rejected: -2.6727
-- Logits/chosen: -2.6886
+- Logits/chosen: -2.6887
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 3.0,
+    "eval_logits/chosen": -2.6887364387512207,
+    "eval_logits/rejected": -2.6727285385131836,
+    "eval_logps/chosen": -75.25064849853516,
+    "eval_logps/rejected": -85.66250610351562,
+    "eval_loss": 0.9378770589828491,
+    "eval_rewards/accuracies": 0.3472222089767456,
+    "eval_rewards/chosen": -0.34173667430877686,
+    "eval_rewards/margins": 1.698632836341858,
+    "eval_rewards/rejected": -2.0403695106506348,
+    "eval_runtime": 119.487,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 16.738,
+    "eval_steps_per_second": 0.527,
     "total_flos": 0.0,
     "train_loss": 0.49774221859604084,
     "train_runtime": 9884.8529,

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

@@ -0,0 +1,16 @@
+{
+    "epoch": 3.0,
+    "eval_logits/chosen": -2.6887364387512207,
+    "eval_logits/rejected": -2.6727285385131836,
+    "eval_logps/chosen": -75.25064849853516,
+    "eval_logps/rejected": -85.66250610351562,
+    "eval_loss": 0.9378770589828491,
+    "eval_rewards/accuracies": 0.3472222089767456,
+    "eval_rewards/chosen": -0.34173667430877686,
+    "eval_rewards/margins": 1.698632836341858,
+    "eval_rewards/rejected": -2.0403695106506348,
+    "eval_runtime": 119.487,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 16.738,
+    "eval_steps_per_second": 0.527
+}

runs/Sep28_00-01-30_qa-a100-005.crc.nd.edu/events.out.tfevents.1727506514.qa-a100-005.crc.nd.edu.5885.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4a2d244415beac609e504e97b3f4d6570a4ec8ac032a5446a05590051ed3f28
+size 828