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-3e-07-0.62-polynomial-3.0
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-align-scan-3e-07-0.62-polynomial-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: 0.8766
-- Rewards/chosen: -0.9523
-- Rewards/rejected: -2.5666
-- Rewards/accuracies: 0.3611
-- Rewards/margins: 1.6143
-- Logps/rejected: -85.2680
-- Logps/chosen: -76.0272
-- Logits/rejected: -2.6367
-- Logits/chosen: -2.6538
+- Loss: 0.8726
+- Rewards/chosen: -0.1874
+- Rewards/rejected: -1.8304
+- Rewards/accuracies: 0.375
+- Rewards/margins: 1.6430
+- Logps/rejected: -84.0806
+- Logps/chosen: -74.7935
+- Logits/rejected: -2.6285
+- Logits/chosen: -2.6453
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 3.0,
+    "eval_logits/chosen": -2.6452701091766357,
+    "eval_logits/rejected": -2.628488779067993,
+    "eval_logps/chosen": -74.79349517822266,
+    "eval_logps/rejected": -84.08062744140625,
+    "eval_loss": 0.8725804686546326,
+    "eval_rewards/accuracies": 0.375,
+    "eval_rewards/chosen": -0.1874038428068161,
+    "eval_rewards/margins": 1.6430120468139648,
+    "eval_rewards/rejected": -1.8304157257080078,
+    "eval_runtime": 113.9064,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 17.558,
+    "eval_steps_per_second": 0.553,
     "total_flos": 0.0,
     "train_loss": 0.4762609164889266,
     "train_runtime": 9670.9508,

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": -2.6452701091766357,
+    "eval_logits/rejected": -2.628488779067993,
+    "eval_logps/chosen": -74.79349517822266,
+    "eval_logps/rejected": -84.08062744140625,
+    "eval_loss": 0.8725804686546326,
+    "eval_rewards/accuracies": 0.375,
+    "eval_rewards/chosen": -0.1874038428068161,
+    "eval_rewards/margins": 1.6430120468139648,
+    "eval_rewards/rejected": -1.8304157257080078,
+    "eval_runtime": 113.9064,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 17.558,
+    "eval_steps_per_second": 0.553
+}

runs/Sep27_13-05-27_qa-a100-005.crc.nd.edu/events.out.tfevents.1727466864.qa-a100-005.crc.nd.edu.4098766.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1dc1e65a072c90d37d2a35e33a78ec111f8f0da619d39da8ab5201a88a405bb
+size 828