16 bit model release?
#3
by
Kernel
- opened
Can you convert please to fp16?
When you download the model you can specify the dtype by doing
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("nomic-ai/gpt4all-13b-snoozy", torch_dtype=torch.half)
zpn
changed discussion status to
closed