Downloads · 30 days
31
9% of all-time downloads
teilomillet_-_Potato-3B-gguf
RichardErkhov/teilomillet_-_Potato-3B-gguf
teilomillet_-_Potato-3B-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.
All-time downloads
336
Public
Repo size
34.7 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.gguf40.2 GB · 100%
At a glance
- Access
- Public
- Created
- Jul 31, 2024
- Updated
- Jul 31, 2024
- SHA
- ea705d3d
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
Potato-3B - GGUF
- Model creator: https://huggingface.co/teilomillet/
- Original model: https://huggingface.co/teilomillet/Potato-3B/
| Name | Quant method | Size |
|---|---|---|
| Potato-3B.Q2_K.gguf | Q2_K | 1.09GB |
| Potato-3B.IQ3_XS.gguf | IQ3_XS | 1.21GB |
| Potato-3B.IQ3_S.gguf | IQ3_S | 1.27GB |
| Potato-3B.Q3_K_S.gguf | Q3_K_S | 1.27GB |
| Potato-3B.IQ3_M.gguf | IQ3_M | 1.33GB |
| Potato-3B.Q3_K.gguf | Q3_K | 1.4GB |
| Potato-3B.Q3_K_M.gguf | Q3_K_M | 1.4GB |
| Potato-3B.Q3_K_L.gguf | Q3_K_L | 1.52GB |
| Potato-3B.IQ4_XS.gguf | IQ4_XS | 1.55GB |
| Potato-3B.Q4_0.gguf | Q4_0 | 1.62GB |
| Potato-3B.IQ4_NL.gguf | IQ4_NL | 1.63GB |
| Potato-3B.Q4_K_S.gguf | Q4_K_S | 1.63GB |
| Potato-3B.Q4_K.gguf | Q4_K | 1.72GB |
| Potato-3B.Q4_K_M.gguf | Q4_K_M | 1.72GB |
| Potato-3B.Q4_1.gguf | Q4_1 | 1.79GB |
| Potato-3B.Q5_0.gguf | Q5_0 | 1.95GB |
| Potato-3B.Q5_K_S.gguf | Q5_K_S | 1.95GB |
| Potato-3B.Q5_K.gguf | Q5_K | 2.01GB |
| Potato-3B.Q5_K_M.gguf | Q5_K_M | 2.01GB |
| Potato-3B.Q5_1.gguf | Q5_1 | 2.12GB |
| Potato-3B.Q6_K.gguf | Q6_K | 2.31GB |
| Potato-3B.Q8_0.gguf | Q8_0 | 2.99GB |
Original model description:
license: apache-2.0 language:
- fr
- en
- zh widget:
- text: "<s> [|User|] Comment faire un bon plat ? </s>[|Assistant|]"
Merging stuff to make a potato. Idk about it, might delete later.
Merge of MiniMerlin via Task arithmetic using mergekit. There was no goal except merging. Interest in the outcome tho. I might need to fine-tune it more.
FT on more french data (Merlin).
Je pense qu'il s'agit du meilleur model français en 3B. Essayez le.
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
model = AutoModelForCausalLM.from_pretrained(
"teilomillet/Potato-3B",
revision="0.1",
return_dict=True,
torch_dtype=torch.bfloat16,
device_map='auto'
)
tokenizer = AutoTokenizer.from_pretrained("teilomillet/Potato-3B")
tokenizer.pad_token = tokenizer.eos_token
text = "[|User|] Comment faire un bon plat ? </s>[|Assistant|]"
inputs = tokenizer(text, return_tensors="pt").to(0)
outputs = model.generate(**inputs, max_new_tokens=800)
print(tokenizer.decode(outputs[0], skip_special_tokens=False))
#merge
- Created
- Jul 31, 2024
- Updated
- Jul 31, 2024