Downloads · 30 days
84
2% of all-time downloads
RichardErkhov/Elizezen_-_Berghof-NSFW-7B-gguf
Elizezen_-_Berghof-NSFW-7B-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
84
2% of all-time downloads
All-time downloads
3.4K
Public
Repo size
82.2 GB
Likes
1
Public
Click a slice to open those files.
.gguf95.2 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
Berghof-NSFW-7B - GGUF
| Name | Quant method | Size |
|---|---|---|
| Berghof-NSFW-7B.Q2_K.gguf | Q2_K | 2.53GB |
| Berghof-NSFW-7B.IQ3_XS.gguf | IQ3_XS | 2.81GB |
| Berghof-NSFW-7B.IQ3_S.gguf | IQ3_S | 2.96GB |
| Berghof-NSFW-7B.Q3_K_S.gguf | Q3_K_S | 2.95GB |
| Berghof-NSFW-7B.IQ3_M.gguf | IQ3_M | 3.06GB |
| Berghof-NSFW-7B.Q3_K.gguf | Q3_K | 3.28GB |
| Berghof-NSFW-7B.Q3_K_M.gguf | Q3_K_M | 3.28GB |
| Berghof-NSFW-7B.Q3_K_L.gguf | Q3_K_L | 3.56GB |
| Berghof-NSFW-7B.IQ4_XS.gguf | IQ4_XS | 3.67GB |
| Berghof-NSFW-7B.Q4_0.gguf | Q4_0 | 3.83GB |
| Berghof-NSFW-7B.IQ4_NL.gguf | IQ4_NL | 3.87GB |
| Berghof-NSFW-7B.Q4_K_S.gguf | Q4_K_S | 3.86GB |
| Berghof-NSFW-7B.Q4_K.gguf | Q4_K | 4.07GB |
| Berghof-NSFW-7B.Q4_K_M.gguf | Q4_K_M | 4.07GB |
| Berghof-NSFW-7B.Q4_1.gguf | Q4_1 | 4.24GB |
| Berghof-NSFW-7B.Q5_0.gguf | Q5_0 | 4.65GB |
| Berghof-NSFW-7B.Q5_K_S.gguf | Q5_K_S | 4.65GB |
| Berghof-NSFW-7B.Q5_K.gguf | Q5_K | 4.78GB |
| Berghof-NSFW-7B.Q5_K_M.gguf | Q5_K_M | 4.78GB |
| Berghof-NSFW-7B.Q5_1.gguf | Q5_1 | 5.07GB |
| Berghof-NSFW-7B.Q6_K.gguf | Q6_K | 5.53GB |
| Berghof-NSFW-7B.Q8_0.gguf | Q8_0 | 7.17GB |
language:
多分これが一番強いと思います
Ensure you are using Transformers 4.34.0 or newer.
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Elizezen/Berghof-NSFW-7B")
model = AutoModelForCausalLM.from_pretrained(
"Elizezen/Berghof-NSFW-7B",
torch_dtype="auto",
)
model.eval()
if torch.cuda.is_available():
model = model.to("cuda")
input_ids = tokenizer.encode(
"吾輩は猫である。名前はまだない",,
add_special_tokens=True,
return_tensors="pt"
)
tokens = model.generate(
input_ids.to(device=model.device),
max_new_tokens=512,
temperature=1,
top_p=0.95,
do_sample=True,
)
out = tokenizer.decode(tokens[0][input_ids.shape[1]:], skip_special_tokens=True).strip()
print(out)
The model is mainly intended to be used for generating novels. It may not be so capable with instruction-based responses.