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: 0.5000
-- Rewards/chosen: -2.1875
-- Rewards/rejected: -3.1436
-- Rewards/accuracies: 0.7320
-- Rewards/margins: 0.9561
-- Logps/rejected: -503.5489
-- Logps/chosen: -403.7713
-- Logits/rejected: -0.0548
-- Logits/chosen: -0.6702
+- Loss: 0.4980
+- Rewards/chosen: -2.1242
+- Rewards/rejected: -3.0843
+- Rewards/accuracies: 0.7380
+- Rewards/margins: 0.9601
+- Logps/rejected: -497.6194
+- Logps/chosen: -397.4371
+- Logits/rejected: -0.2690
+- Logits/chosen: -0.8689
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9997382884061764,
+    "eval_logits/chosen": -0.8689022064208984,
+    "eval_logits/rejected": -0.2690173089504242,
+    "eval_logps/chosen": -397.43707275390625,
+    "eval_logps/rejected": -497.619384765625,
+    "eval_loss": 0.49802955985069275,
+    "eval_rewards/accuracies": 0.7379999756813049,
+    "eval_rewards/chosen": -2.1241602897644043,
+    "eval_rewards/margins": 0.9601246118545532,
+    "eval_rewards/rejected": -3.084285020828247,
+    "eval_runtime": 316.8163,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 6.313,
+    "eval_steps_per_second": 0.789,
     "total_flos": 0.0,
     "train_loss": 0.5374546366836388,
     "train_runtime": 38043.1849,

config.json

@@ -21,6 +21,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.40.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9997382884061764,
+    "eval_logits/chosen": -0.8689022064208984,
+    "eval_logits/rejected": -0.2690173089504242,
+    "eval_logps/chosen": -397.43707275390625,
+    "eval_logps/rejected": -497.619384765625,
+    "eval_loss": 0.49802955985069275,
+    "eval_rewards/accuracies": 0.7379999756813049,
+    "eval_rewards/chosen": -2.1241602897644043,
+    "eval_rewards/margins": 0.9601246118545532,
+    "eval_rewards/rejected": -3.084285020828247,
+    "eval_runtime": 316.8163,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 6.313,
+    "eval_steps_per_second": 0.789
+}

runs/Apr27_16-08-02_pluslab01-a100.cs.ucla.edu/events.out.tfevents.1714298185.pluslab01-a100.cs.ucla.edu.4087754.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:996211101d24c47d1e76b115b15ecdf0cfd646798ca4ca23925686c1f0ffb626
+size 828