Improving coding ability from
FelixChao/WizardDolphin-7B.
1models:
2 - model: FelixChao/WizardDolphin-7B
3 - model: beowolx/CodeNinja-1.0-OpenChat-7B
4 parameters:
5 density: 0.53
6 weight: 0.3
7 - model: beowolx/MistralHermes-CodePro-7B-v1
8 parameters:
9 density: 0.53
10 weight: 0.3
11merge_method: dare_ties
12base_model: FelixChao/WizardDolphin-7B
13parameters:
14 int8_mask: true
15dtype: bfloat16
1!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "FelixChao/NinjaDolphin-7B"
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"])
Detailed results can be found
here
