Initial model upload

README.md

@@ -1,7 +1,7 @@
 ---
-base_model: koshirowada/pythia_14m_sft
+base_model: koshirowada/pythia_70m_sft
 library_name: transformers
-model_name: pythia_14m_dpo
+model_name: pythia_70m_dpo
 tags:
 - generated_from_trainer
 - trl
@@ -9,9 +9,9 @@ tags:
 licence: license
 ---
 
-# Model Card for pythia_14m_dpo
+# Model Card for pythia_70m_dpo
 
-This model is a fine-tuned version of [koshirowada/pythia_14m_sft](https://huggingface.co/koshirowada/pythia_14m_sft).
+This model is a fine-tuned version of [koshirowada/pythia_70m_sft](https://huggingface.co/koshirowada/pythia_70m_sft).
 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="koshirowada/pythia_14m_dpo", device="cuda")
+generator = pipeline("text-generation", model="koshirowada/pythia_70m_dpo", 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/kosiro131219/pythia_dpo/runs/9gkiulr6)
+[<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/kosiro131219/pythia_dpo/runs/85ieyd7e)
 
 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).