KeyError: 'stripedhyena'

#1
by FumaNet - opened

Hello,
I was trying to run this model with minimal code, using the 1.1_fix revision as suggested, using the following code:

from transformers import AutoConfig, AutoModelForCausalLM

model_name = 'togethercomputer/evo-1-8k-base'

config = AutoConfig.from_pretrained(model_name, trust_remote_code=True, revision="1.1_fix")
model = AutoModelForCausalLM.from_pretrained(
model_name,
config=config,
trust_remote_code=True,
revision="1.1_fix"
)

and run into the following error:

Traceback (most recent call last):
File "hf_example_inference.py", line 5, in
config = AutoConfig.from_pretrained(model_name, trust_remote_code=True, revision="1.1_fix")
...
in getitem
raise KeyError(key)
KeyError: 'stripedhyena'

I am working on Ubuntu:
evo-model==0.4
transformers==4.11.3
flash-attn==2.7.0.post2
torch==2.4.1

Thanks for the help!

Turns out I just had to upgrade transformers, solved now. My bad ^^

FumaNet changed discussion status to closed

Sign up or log in to comment