Model save

README.md

@@ -2,13 +2,9 @@
 license: apache-2.0
 base_model: mistralai/Mistral-7B-Instruct-v0.2
 tags:
-- alignment-handbook
-- generated_from_trainer
 - trl
 - dpo
 - generated_from_trainer
-datasets:
-- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: 0.001_withdpo_4iters_bs256_5102lr_misit_iter_1
   results: []
@@ -19,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # 0.001_withdpo_4iters_bs256_5102lr_misit_iter_1
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the HuggingFaceH4/ultrafeedback_binarized dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the None dataset.
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.99,
     "train_loss": 0.5465719497809975,
-    "train_runtime": 1934.9087,
+    "train_runtime": 1933.9078,
     "train_samples": 15283,
-    "train_samples_per_second": 7.899,
-    "train_steps_per_second": 0.03
+    "train_samples_per_second": 7.903,
+    "train_steps_per_second": 0.031
 }

config.json

@@ -21,6 +21,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.36.2",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 32000
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.99,
     "train_loss": 0.5465719497809975,
-    "train_runtime": 1934.9087,
+    "train_runtime": 1933.9078,
     "train_samples": 15283,
-    "train_samples_per_second": 7.899,
-    "train_steps_per_second": 0.03
+    "train_samples_per_second": 7.903,
+    "train_steps_per_second": 0.031
 }

trainer_state.json

@@ -97,9 +97,9 @@
       "step": 59,
       "total_flos": 0.0,
       "train_loss": 0.5465719497809975,
-      "train_runtime": 1934.9087,
-      "train_samples_per_second": 7.899,
-      "train_steps_per_second": 0.03
+      "train_runtime": 1933.9078,
+      "train_samples_per_second": 7.903,
+      "train_steps_per_second": 0.031
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6871fa8a03cb4d5a33c0905dffdae20d296f7ca28320e447c58912d7f13c95fd
+oid sha256:a2a59ffef973459b6e57008bbd09a476236def5a63e68369ebef0b378a5721f2
 size 6072