README.md

metadata

license: cc-by-nc-nd-4.0
language:
  - en
base_model: vandijklab/C2S-Scale-Pythia-1b-pt
library_name: transformers
tags:
  - biology
  - scRNAseq
  - mlx
  - mlx-my-repo

CD4cell/C2S-Scale-Pythia-1b-pt-mlx-fp16

The Model CD4cell/C2S-Scale-Pythia-1b-pt-mlx-fp16 was converted to MLX format from vandijklab/C2S-Scale-Pythia-1b-pt using mlx-lm version 0.22.3.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("CD4cell/C2S-Scale-Pythia-1b-pt-mlx-fp16")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)