Views
No views yet
| Model | AGIEval | GPT4All | TruthfulQA | Bigbench | Average |
|---|---|---|---|---|---|
| Meta-Llama-3-12B-Instruct | 41.7 | 67.71 | 52.75 | 40.58 | 50.69 |
| Meta-Llama-3-12B | 29.46 | 68.01 | 41.02 | 35.57 | 43.52 |
1slices:
2 - sources:
3 - model: NousResearch/Meta-Llama-3-8B-Instruct
4 layer_range: [0,9]
5 - sources:
6 - model: NousResearch/Meta-Llama-3-8B-Instruct
7 layer_range: [5,14]
8 - sources:
9 - model: NousResearch/Meta-Llama-3-8B-Instruct
10 layer_range: [10,19]
11 - sources:
12 - model: NousResearch/Meta-Llama-3-8B-Instruct
13 layer_range: [15,24]
14 - sources:
15 - model: NousResearch/Meta-Llama-3-8B-Instruct
16 layer_range: [20,32]
17merge_method: passthrough
18dtype: bfloat161!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "mlabonne/Meta-Llama-3-12B-Instruct"
8messages = [{"role": "user", "content": "What is a large language model?"}]
9
10tokenizer = AutoTokenizer.from_pretrained(model)
11prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
12pipeline = transformers.pipeline(
13 "text-generation",
14 model=model,
15 torch_dtype=torch.float16,
16 device_map="auto",
17)
18
19outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
20print(outputs[0]["generated_text"])