Views
No views yet
1dtype: float16
2merge_method: passthrough
3slices:
4- sources:
5 - layer_range: [0, 8]
6 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
7- sources:
8 - layer_range: [4, 12]
9 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
10- sources:
11 - layer_range: [8, 16]
12 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
13- sources:
14 - layer_range: [12, 20]
15 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
16- sources:
17 - layer_range: [16, 24]
18 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
19- sources:
20 - layer_range: [20, 28]
21 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
22- sources:
23 - layer_range: [24, 32]
24 model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp1!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "mvpmaster/Einstein-4D-Marcoro14-12b-32k-experiment"
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"])