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-2e-07-0.39-polynomial-1.0
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-align-scan-2e-07-0.39-polynomial-1.0
 
-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: 0.6324
-- Rewards/chosen: 0.2521
-- Rewards/rejected: -0.1128
+- Loss: 0.6326
+- Rewards/chosen: 0.3567
+- Rewards/rejected: -0.0406
 - Rewards/accuracies: 0.3492
-- Rewards/margins: 0.3648
-- Logps/rejected: -81.4175
-- Logps/chosen: -73.8449
-- Logits/rejected: -2.5055
-- Logits/chosen: -2.5222
+- Rewards/margins: 0.3973
+- Logps/rejected: -81.2324
+- Logps/chosen: -73.5766
+- Logits/rejected: -2.5151
+- Logits/chosen: -2.5314
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.531444787979126,
+    "eval_logits/rejected": -2.5150935649871826,
+    "eval_logps/chosen": -73.5765609741211,
+    "eval_logps/rejected": -81.2323989868164,
+    "eval_loss": 0.6325609087944031,
+    "eval_rewards/accuracies": 0.3492063581943512,
+    "eval_rewards/chosen": 0.3567197322845459,
+    "eval_rewards/margins": 0.39729830622673035,
+    "eval_rewards/rejected": -0.04057859256863594,
+    "eval_runtime": 114.1594,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 17.519,
+    "eval_steps_per_second": 0.552,
     "total_flos": 0.0,
     "train_loss": 0.6390685353960309,
     "train_runtime": 3341.9408,

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": 1.0,
+    "eval_logits/chosen": -2.531444787979126,
+    "eval_logits/rejected": -2.5150935649871826,
+    "eval_logps/chosen": -73.5765609741211,
+    "eval_logps/rejected": -81.2323989868164,
+    "eval_loss": 0.6325609087944031,
+    "eval_rewards/accuracies": 0.3492063581943512,
+    "eval_rewards/chosen": 0.3567197322845459,
+    "eval_rewards/margins": 0.39729830622673035,
+    "eval_rewards/rejected": -0.04057859256863594,
+    "eval_runtime": 114.1594,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 17.519,
+    "eval_steps_per_second": 0.552
+}

runs/Sep27_20-59-21_qa-a100-005.crc.nd.edu/events.out.tfevents.1727488991.qa-a100-005.crc.nd.edu.4172210.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:337b12dfe781d84f02ad8cf47277cb1d1ae2ad6a42d9445cff31dc8ec6bc5402
+size 828