Model save

README.md

@@ -1,7 +1,7 @@
 ---
-base_model: RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter2
+base_model: RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7
 library_name: transformers
-model_name: self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7
+model_name: self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter4_lr1e-7
 tags:
 - generated_from_trainer
 - trl
@@ -9,9 +9,9 @@ tags:
 licence: license
 ---
 
-# Model Card for self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7
+# Model Card for self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter4_lr1e-7
 
-This model is a fine-tuned version of [RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter2](https://huggingface.co/RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter2).
+This model is a fine-tuned version of [RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7](https://huggingface.co/RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -20,14 +20,14 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 from transformers import pipeline
 
 question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7", device="cuda")
+generator = pipeline("text-generation", model="RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter4_lr1e-7", device="cuda")
 output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
 print(output["generated_text"])
 ```
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yyr/huggingface/runs/gdrfivg0)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yyr/huggingface/runs/md0w7ied)
 
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
 

last_checkpoint/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter2",
+  "_name_or_path": "RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter3_lr1e-7",
   "architectures": [
     "MistralForCausalLM"
   ],

last_checkpoint/model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b63950ac26992f37b69fac54d7257fdf06f817db5c7912ebda1581cddbeeda60
+oid sha256:a307e383b148cd1e63ea4f206644e5d2930a4bfa0e2192a4bdfdb6ec7ee9e5cd
 size 4983016096

last_checkpoint/model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:478b4753d75a532db9d1811e86b9fec4137f72540c9397e1f8eb86e696a4b945
+oid sha256:0e41928d9aba5bd1e1fe9f469329e2c59603c6f268c4ed9dae8157a8790d722c
 size 4999836776

last_checkpoint/model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89ce15203f434e9a1b6da0065e8cc1057369f77e5d34d4871b85c95dd7fbbbc1
+oid sha256:3670140d8801a73d40fa130e15adf4c5cc45830b5d844ef71ae16ca8e6b4f6e5
 size 4983067960

last_checkpoint/model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a1c3cbdaeed31d9ec1743500409cb0f311917dbcb15c83b80e67dff94c172a9
+oid sha256:b6ede8c313fefb7942fb139d0e853b9f00116f8a5e2d7f945e1266c8bf07ba57
 size 1073750144

last_checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0961bc15186ae81c5777e901bef7d15ec9c87865609f5706bc520911646c0e74
+oid sha256:88af4b6c33d5b8a38584956dd23c9d0fb2d2bde238277ce3724b976369b9b088
 size 8056