Views
No views yet
1slices:
2 - sources:
3 - model: yam-peleg/Experiment26-7B
4 layer_range: [0, 5]
5 - sources:
6 - model: yam-peleg/Experiment26-7B
7 layer_range: [3, 8]
8 - sources:
9 - model: yam-peleg/Experiment26-7B
10 layer_range: [6, 11]
11 - sources:
12 - model: yam-peleg/Experiment26-7B
13 layer_range: [9, 14]
14 - sources:
15 - model: yam-peleg/Experiment26-7B
16 layer_range: [12, 17]
17 - sources:
18 - model: yam-peleg/Experiment26-7B
19 layer_range: [15, 20]
20 - sources:
21 - model: yam-peleg/Experiment26-7B
22 layer_range: [18, 23]
23 - sources:
24 - model: yam-peleg/Experiment26-7B
25 layer_range: [21, 26]
26 - sources:
27 - model: yam-peleg/Experiment26-7B
28 layer_range: [24, 29]
29 - sources:
30 - model: yam-peleg/Experiment26-7B
31 layer_range: [27, 32]
32merge_method: passthrough
33dtype: bfloat161!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "babybirdprd/Experiment26-7B-passthrough-10slice"
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"])