Downloads · 30 days
71
16% of all-time downloads
flemmingmiguel_-_MBX-7B-v3-gguf
RichardErkhov/flemmingmiguel_-_MBX-7B-v3-gguf
flemmingmiguel_-_MBX-7B-v3-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
437
Public
Repo size
7.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf7.2 GB · 100%
At a glance
- Access
- Public
- Created
- Jun 26, 2024
- Updated
- Jun 26, 2024
- SHA
- 31770386
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
MBX-7B-v3 - GGUF
- Model creator: https://huggingface.co/flemmingmiguel/
- Original model: https://huggingface.co/flemmingmiguel/MBX-7B-v3/
| Name | Quant method | Size |
|---|---|---|
| MBX-7B-v3.Q2_K.gguf | Q2_K | 2.53GB |
| MBX-7B-v3.IQ3_XS.gguf | IQ3_XS | 2.81GB |
| MBX-7B-v3.IQ3_S.gguf | IQ3_S | 1.12GB |
| MBX-7B-v3.Q3_K_S.gguf | Q3_K_S | 0.21GB |
| MBX-7B-v3.IQ3_M.gguf | IQ3_M | 0.05GB |
| MBX-7B-v3.Q3_K.gguf | Q3_K | 0.01GB |
| MBX-7B-v3.Q3_K_M.gguf | Q3_K_M | 0.01GB |
| MBX-7B-v3.Q3_K_L.gguf | Q3_K_L | 0.0GB |
| MBX-7B-v3.IQ4_XS.gguf | IQ4_XS | 0.0GB |
| MBX-7B-v3.Q4_0.gguf | Q4_0 | 0.0GB |
| MBX-7B-v3.IQ4_NL.gguf | IQ4_NL | 0.0GB |
| MBX-7B-v3.Q4_K_S.gguf | Q4_K_S | 0.0GB |
| MBX-7B-v3.Q4_K.gguf | Q4_K | 0.0GB |
| MBX-7B-v3.Q4_K_M.gguf | Q4_K_M | 0.0GB |
| MBX-7B-v3.Q4_1.gguf | Q4_1 | 0.0GB |
| MBX-7B-v3.Q5_0.gguf | Q5_0 | 0.0GB |
| MBX-7B-v3.Q5_K_S.gguf | Q5_K_S | 0.0GB |
| MBX-7B-v3.Q5_K.gguf | Q5_K | 0.0GB |
| MBX-7B-v3.Q5_K_M.gguf | Q5_K_M | 0.0GB |
| MBX-7B-v3.Q5_1.gguf | Q5_1 | 0.0GB |
| MBX-7B-v3.Q6_K.gguf | Q6_K | 0.0GB |
| MBX-7B-v3.Q8_0.gguf | Q8_0 | 0.0GB |
Original model description:
license: apache-2.0 tags:
- merge
- mergekit
- lazymergekit
- flemmingmiguel/MBX-7B
- flemmingmiguel/MBX-7B-v3
MBX-7B-v3
MBX-7B-v3 is a merge of the following models using LazyMergekit:
🧩 Configuration
slices:
- sources:
- model: flemmingmiguel/MBX-7B
layer_range: [0, 32]
- model: flemmingmiguel/MBX-7B-v3
layer_range: [0, 32]
merge_method: slerp
base_model: flemmingmiguel/MBX-7B-v3
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.45 # fallback for rest of tensors
dtype: float16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "flemmingmiguel/MBX-7B-v3"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
- Created
- Jun 26, 2024
- Updated
- Jun 26, 2024