Update README.md

README.md

@@ -1,4 +1,14 @@
-Sophie0-SFT
+---
+license: apache-2.0
+datasets:
+- K-and-K/knights-and-knaves
+language:
+- en
+- zh
+base_model:
+- SophieA17/Sophie0-Reasoning-SFT
+---
+Sophie0-Reasoning-GRPO
 
 ### Introduction
 
@@ -18,8 +28,8 @@ import torch
 
 from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
 
-model: AutoModelForCausalLM = AutoModelForCausalLM.from_pretrained("SophieA17/Sophie0-SFT", trust_remote_code=True)
-tokenizer: AutoTokenizer = AutoTokenizer.from_pretrained("SophieA17/Sophie0-SFT", trust_remote_code=True)
+model: AutoModelForCausalLM = AutoModelForCausalLM.from_pretrained("SophieA17/Sophie0-Reasoning-GRPO", trust_remote_code=True)
+tokenizer: AutoTokenizer = AutoTokenizer.from_pretrained("SophieA17/Sophie0-Reasoning-GRPO", trust_remote_code=True)
 
 model = model.to(device="cuda:0", dtype=torch.bfloat16)
 inputs = [