PRIME-RL
/

EurusPRM-Stage1

Model card Files Files and versions

yuchenFan commited on Jan 2, 2025

Commit

7dda72b

·

1 Parent(s): 06334c3

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -72,8 +72,8 @@ We show an example leveraging **EurusPRM-Stage1** below:
 coef=0.001
 d = {'query':'111','answer':['111','222']
      }
-model = AutoModelForCausalLM.from_pretrained('PRIME-RL/EurusPRM-Stage2')
-tokenizer = AutoTokenizer.from_pretrained('PRIME-RL/EurusPRM-Stage2')
 ref_model = AutoModelForCausalLM.from_pretrained('Qwen/Qwen2.5-Math-7B-Instruct')
 input_ids = tokenizer.apply_chat_template([
     {"role": "user", "content": d["query"]},

 coef=0.001
 d = {'query':'111','answer':['111','222']
      }
+model = AutoModelForCausalLM.from_pretrained('PRIME-RL/EurusPRM-Stage1')
+tokenizer = AutoTokenizer.from_pretrained('PRIME-RL/EurusPRM-Stage1')
 ref_model = AutoModelForCausalLM.from_pretrained('Qwen/Qwen2.5-Math-7B-Instruct')
 input_ids = tokenizer.apply_chat_template([
     {"role": "user", "content": d["query"]},