Views
No views yet
llama3 codebase.generate() function. Let's see examples of both.1import transformers
2import torch
3
4model_id = "meta-llama/Meta-Llama-3-8B-Instruct"
5
6pipeline = transformers.pipeline(
7 "text-generation",
8 model=model_id,
9 model_kwargs={"torch_dtype": torch.bfloat16},
10 device_map="auto",
11)
12
13messages = [
14 {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
15 {"role": "user", "content": "Who are you?"},
16]
17
18prompt = pipeline.tokenizer.apply_chat_template(
19 messages,
20 tokenize=False,
21 add_generation_prompt=True
22)
23
24terminators = [
25 pipeline.tokenizer.eos_token_id,
26 pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
27]
28
29outputs = pipeline(
30 prompt,
31 max_new_tokens=256,
32 eos_token_id=terminators,
33 do_sample=True,
34 temperature=0.6,
35 top_p=0.9,
36)
37print(outputs[0]["generated_text"][len(prompt):])1from transformers import AutoTokenizer, AutoModelForCausalLM
2import torch
3
4model_id = "meta-llama/Meta-Llama-3-8B-Instruct"
5
6tokenizer = AutoTokenizer.from_pretrained(model_id)
7model = AutoModelForCausalLM.from_pretrained(
8 model_id,
9 torch_dtype=torch.bfloat16,
10 device_map="auto",
11)
12
13messages = [
14 {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
15 {"role": "user", "content": "Who are you?"},
16]
17
18input_ids = tokenizer.apply_chat_template(
19 messages,
20 add_generation_prompt=True,
21 return_tensors="pt"
22).to(model.device)
23
24terminators = [
25 tokenizer.eos_token_id,
26 tokenizer.convert_tokens_to_ids("<|eot_id|>")
27]
28
29outputs = model.generate(
30 input_ids,
31 max_new_tokens=256,
32 eos_token_id=terminators,
33 do_sample=True,
34 temperature=0.6,
35 top_p=0.9,
36)
37response = outputs[0][input_ids.shape[-1]:]
38print(tokenizer.decode(response, skip_special_tokens=True))