Downloads · 30 days
795
52% of all-time downloads
sudy-super/baku-13b
baku-13b is a text generation model from sudy-super. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This model is a 13.5 billion parameter model that combines four sets of 16 layers each from CALM2-7B.
Downloads · 30 days
795
52% of all-time downloads
All-time downloads
1.5K
Public
Repo size
53.9 GB
Likes
1
Public
Click a slice to open those files.
.bin27 GB · 100%
From the Hugging Face model README
This model is a 13.5 billion parameter model that combines four sets of 16 layers each from CALM2-7B.
This model is experimental and may not achieve expected performance without additional tuning.
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
tokenizer = AutoTokenizer.from_pretrained("sudy-super/baku-13b")
model = AutoModelForCausalLM.from_pretrained("sudy-super/baku-13b", device_map="auto", torch_dtype=torch.bfloat16)
prompt = "大規模言語モデルとは、"
token_ids = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt")
with torch.no_grad():
output_ids = model.generate(
token_ids.to(model.device),
max_new_tokens=100,
do_sample=True,
temperature=0.8,
pad_token_id=tokenizer.pad_token_id,
bos_token_id=tokenizer.bos_token_id,
eos_token_id=tokenizer.eos_token_id
)
result = tokenizer.decode(output_ids.tolist()[0])
print(result)