Downloads · 30 days
141
7% of all-time downloads
Muhammad2003_-_OrpoLlama3-8B-gguf
RichardErkhov/Muhammad2003_-_OrpoLlama3-8B-gguf
Muhammad2003_-_OrpoLlama3-8B-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
2.1K
Public
Repo size
92.9 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf108 GB · 100%
At a glance
- Access
- Public
- Created
- Jul 19, 2024
- Updated
- Jul 20, 2024
- SHA
- fd316abf
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
OrpoLlama3-8B - GGUF
- Model creator: https://huggingface.co/Muhammad2003/
- Original model: https://huggingface.co/Muhammad2003/OrpoLlama3-8B/
| Name | Quant method | Size |
|---|---|---|
| OrpoLlama3-8B.Q2_K.gguf | Q2_K | 2.96GB |
| OrpoLlama3-8B.IQ3_XS.gguf | IQ3_XS | 3.28GB |
| OrpoLlama3-8B.IQ3_S.gguf | IQ3_S | 3.43GB |
| OrpoLlama3-8B.Q3_K_S.gguf | Q3_K_S | 3.41GB |
| OrpoLlama3-8B.IQ3_M.gguf | IQ3_M | 3.52GB |
| OrpoLlama3-8B.Q3_K.gguf | Q3_K | 3.74GB |
| OrpoLlama3-8B.Q3_K_M.gguf | Q3_K_M | 3.74GB |
| OrpoLlama3-8B.Q3_K_L.gguf | Q3_K_L | 4.03GB |
| OrpoLlama3-8B.IQ4_XS.gguf | IQ4_XS | 4.18GB |
| OrpoLlama3-8B.Q4_0.gguf | Q4_0 | 4.34GB |
| OrpoLlama3-8B.IQ4_NL.gguf | IQ4_NL | 4.38GB |
| OrpoLlama3-8B.Q4_K_S.gguf | Q4_K_S | 4.37GB |
| OrpoLlama3-8B.Q4_K.gguf | Q4_K | 4.58GB |
| OrpoLlama3-8B.Q4_K_M.gguf | Q4_K_M | 4.58GB |
| OrpoLlama3-8B.Q4_1.gguf | Q4_1 | 4.78GB |
| OrpoLlama3-8B.Q5_0.gguf | Q5_0 | 5.21GB |
| OrpoLlama3-8B.Q5_K_S.gguf | Q5_K_S | 5.21GB |
| OrpoLlama3-8B.Q5_K.gguf | Q5_K | 5.34GB |
| OrpoLlama3-8B.Q5_K_M.gguf | Q5_K_M | 5.34GB |
| OrpoLlama3-8B.Q5_1.gguf | Q5_1 | 5.65GB |
| OrpoLlama3-8B.Q6_K.gguf | Q6_K | 6.14GB |
| OrpoLlama3-8B.Q8_0.gguf | Q8_0 | 7.95GB |
Original model description:
language:
- en license: apache-2.0 library_name: transformers tags:
- orpo
base_model:
- meta-llama/Meta-Llama-3-8B datasets:
- mlabonne/orpo-dpo-mix-40k
OrpoLlama3-8B
This is an ORPO fine-tune of meta-llama/Meta-Llama-3-8B on 15k steps of mlabonne/orpo-dpo-mix-40k.
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Muhammad2003/OrpoLlama3-8B"
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"])
📈 Training curves
Wandb Report
🏆 Evaluation
- Created
- Jul 19, 2024
- Updated
- Jul 20, 2024