mlx-community
/

Llama-2-7b-chat-4-bit

Text Generation

Model card Files Files and versions Community

Llama-2-7b-chat-4-bit / config.json

reach-vb's picture

reach-vb HF staff

Upload config.json with huggingface_hub

6921e0e 6 months ago

raw history blame

No virus

227 Bytes

	{
	"dim": 4096,
	"multiple_of": 256,
	"n_heads": 32,
	"n_layers": 32,
	"norm_eps": 1e-05,
	"vocab_size": -1,
	"model_type": "llama",
	"quantization": {
	"group_size": 64,
	"bits": 4
	}
	}