Model save

README.md

@@ -1,7 +1,7 @@
 ---
-base_model: RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter4
+base_model: RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter6
 library_name: transformers
-model_name: self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter5_lr3e-7
+model_name: self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter7
 tags:
 - generated_from_trainer
 - trl
@@ -9,9 +9,9 @@ tags:
 licence: license
 ---
 
-# Model Card for self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter5_lr3e-7
+# Model Card for self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter7
 
-This model is a fine-tuned version of [RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter4](https://huggingface.co/RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter4).
+This model is a fine-tuned version of [RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter6](https://huggingface.co/RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter6).
 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_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter5_lr3e-7", device="cuda")
+generator = pipeline("text-generation", model="RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter7", 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/ylxdw82v)
+[<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/93v2aobh)
 
 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_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter4",
+  "_name_or_path": "RyanYr/self-correct_Llama-3.2-3B-Instruct_metaMathQA_dpo_iter6",
   "architectures": [
     "LlamaForCausalLM"
   ],

last_checkpoint/model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d61a88f70807d4b8b2e87163f02bb586512e415085f262d8e094642bd56b9148
+oid sha256:8b3743fdd0699116d5d08617f94b43ffd125982f766ca04f43b378c9a84473fb
 size 4965805240

last_checkpoint/model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8d100ac996183a3c659a7d1044b24ba1a2eace568d936d97f2bc45c9c68b57c
+oid sha256:a1c1aea868f78cbdd62d3aa23fe8ee0dbb97432e43523d85f9e2af05e670c228
 size 2247741136

last_checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18dbcf515a5ee5216736b959d1516b6ddc7b4530e2cff9b5ed5530752254ea1f
+oid sha256:7e9ad152a9b6faee28771ed34b7f3e4b7af1a707913b6015acf419ab8976cd59
 size 7608