Text Generation
Transformers
Safetensors
French
English
llama
legal
code
text-generation-inference
art
conversational
Inference Endpoints
manu commited on
Commit
46d96f8
1 Parent(s): 9cef06a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -24,7 +24,7 @@ This model is part of the CroissantLLM initiative, and corresponds to the checkp
24
 
25
  https://arxiv.org/abs/2402.00786
26
 
27
- For best performance, it should be used with a temperature of above 0.4, and with the exact template described below:
28
 
29
  ```python
30
  chat = [
@@ -85,7 +85,7 @@ chat = [
85
  chat_input = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
86
 
87
  inputs = tokenizer(chat_input, return_tensors="pt", add_special_tokens=True).to(model.device)
88
- tokens = model.generate(**inputs, max_new_tokens=150, do_sample=True, top_p=0.95, top_k=60, temperature=0.5)
89
  print(tokenizer.decode(tokens[0]))
90
  ```
91
 
 
24
 
25
  https://arxiv.org/abs/2402.00786
26
 
27
+ For best performance, it should be used with a temperature of 0.3 or more, and with the exact template described below:
28
 
29
  ```python
30
  chat = [
 
85
  chat_input = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
86
 
87
  inputs = tokenizer(chat_input, return_tensors="pt", add_special_tokens=True).to(model.device)
88
+ tokens = model.generate(**inputs, max_new_tokens=150, do_sample=True, top_p=0.95, top_k=60, temperature=0.3)
89
  print(tokenizer.decode(tokens[0]))
90
  ```
91