simsim314's picture
Update README.md
4d8d64c
|
raw
history blame
363 Bytes
metadata
license: gpl

Sharded version of This model. Use the tokenizer from there

from transformers import LlamaTokenizer, AutoModelForCausalLM

tokenizer = LlamaTokenizer.from_pretrained("NousResearch/Nous-Hermes-13b")
model = AutoModelForCausalLM.from_pretrained("simsim314/Hermes-13b-hf-shards")