Model save

README.md

@@ -3,16 +3,10 @@ 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
   results: []
@@ -23,17 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-align-scan
 
-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: 0.6700
-- Rewards/chosen: 0.0017
-- Rewards/rejected: -0.0506
-- Rewards/accuracies: 0.3155
-- Rewards/margins: 0.0522
-- Logps/rejected: -86.1855
-- Logps/chosen: -74.3248
-- Logits/rejected: -2.4181
-- Logits/chosen: -2.4359
+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.
 
 ## Model description
 

all_results.json

@@ -15,8 +15,8 @@
     "eval_steps_per_second": 0.544,
     "total_flos": 0.0,
     "train_loss": 0.0,
-    "train_runtime": 0.0191,
+    "train_runtime": 0.0422,
     "train_samples": 6113,
-    "train_samples_per_second": 319809.664,
-    "train_steps_per_second": 5022.367
+    "train_samples_per_second": 144824.787,
+    "train_steps_per_second": 2274.363
 }

config.json

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

train_results.json

@@ -2,8 +2,8 @@
     "epoch": 1.0,
     "total_flos": 0.0,
     "train_loss": 0.0,
-    "train_runtime": 0.0191,
+    "train_runtime": 0.0422,
     "train_samples": 6113,
-    "train_samples_per_second": 319809.664,
-    "train_steps_per_second": 5022.367
+    "train_samples_per_second": 144824.787,
+    "train_steps_per_second": 2274.363
 }

trainer_state.json

@@ -163,9 +163,9 @@
       "step": 96,
       "total_flos": 0.0,
       "train_loss": 0.0,
-      "train_runtime": 0.0191,
-      "train_samples_per_second": 319809.664,
-      "train_steps_per_second": 5022.367
+      "train_runtime": 0.0422,
+      "train_samples_per_second": 144824.787,
+      "train_steps_per_second": 2274.363
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7a254b52f3726f429bed8a814c369a708a14aeeaffe6d431c6d25fb5b741d60
+oid sha256:c7376d9c3dca4c85a08d582603a55b02eadbd6d47a53baa367c70ada442d7557
 size 7544