Views
No views yet
Ananda100/pocketcoder100M-distilled
on a custom Python coding-problem dataset, using the deepseek-ai/deepseek-coder-6.7b-base tokenizer.transformers model, so AutoModel.from_pretrained
won't work out of the box:1import json, torch
2from safetensors.torch import load_file
3
4with open("config.json") as f:
5 cfg = json.load(f)
6
7config = GPTConfig(**cfg)
8model = GPT(config)
9state_dict = load_file("model.safetensors")
10model.load_state_dict(state_dict)
11model.eval()