Bio-Mistralv2-Squared is a merge of pre-trained language models created using
mergekit.
This model was merged using the SLERP merge method.
1
2slices:
3 - sources:
4 - model: BioMistral/BioMistral-7B
5 layer_range: [0, 32]
6 - model: mistralai/Mistral-7B-Instruct-v0.2
7 layer_range: [0, 32]
8merge_method: slerp
9base_model: BioMistral/BioMistral-7B
10parameters:
11 t:
12 - filter: self_attn
13 value: [0, 0.5, 0.3, 0.7, 1]
14 - filter: mlp
15 value: [1, 0.5, 0.7, 0.3, 0]
16 - value: 0.5
17dtype: bfloat16
18
1!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "Kabster/Bio-Mistralv2-Squared"
8messages = [{"role": "user", "content": "What is fluimucil used for?"}]
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=0.2, top_k=100, top_p=0.95)
20print(outputs[0]["generated_text"])
