mistral-inference icon indicating copy to clipboard operation
mistral-inference copied to clipboard

Missing the params.json

Open littlewwwhite opened this issue 1 year ago • 3 comments

FileNotFoundError: [Errno 2] No such file or directory: 'Mistral-7B-Instruct-v0.2/params.json'

and I replace the params.json to config.json but still lack of several params such like kv_cache dim etc.

littlewwwhite avatar Mar 19 '24 14:03 littlewwwhite

Same here. Also, the model on huggingface has different names for w1/w2/w3 weights (up_proj, down_proj, gate_proj), and it's not documented anywhere. I guess the model from HF works with the Transformers library, but both this and MLX impl need fixing to use it.

kolinko avatar Apr 10 '24 06:04 kolinko

Just add the different ones to your params.json file for w1/w2/w3 etc…

nathan-pugh-fsn avatar Apr 11 '24 23:04 nathan-pugh-fsn

When you say different ones, you mean just add a bunch of random weights? Are these normalized between 0 and 1? By the way, this thread seems to offer at least a proposed set of params, if anyone gets here and wants an answer: https://github.com/vikhyat/mixtral-inference/issues/3

mmoerman avatar Jul 16 '24 12:07 mmoerman