Downloads · 30 days
41
15% of all-time downloads
uproai_-_Rose-2x7B-gguf
RichardErkhov/uproai_-_Rose-2x7B-gguf
uproai_-_Rose-2x7B-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
281
Public
Repo size
146 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf169 GB · 100%
At a glance
- Access
- Public
- Created
- Sep 22, 2024
- Updated
- Sep 22, 2024
- SHA
- 86b17a47
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
Rose-2x7B - GGUF
- Model creator: https://huggingface.co/uproai/
- Original model: https://huggingface.co/uproai/Rose-2x7B/
| Name | Quant method | Size |
|---|---|---|
| Rose-2x7B.Q2_K.gguf | Q2_K | 4.43GB |
| Rose-2x7B.IQ3_XS.gguf | IQ3_XS | 4.95GB |
| Rose-2x7B.IQ3_S.gguf | IQ3_S | 5.22GB |
| Rose-2x7B.Q3_K_S.gguf | Q3_K_S | 5.2GB |
| Rose-2x7B.IQ3_M.gguf | IQ3_M | 5.35GB |
| Rose-2x7B.Q3_K.gguf | Q3_K | 5.78GB |
| Rose-2x7B.Q3_K_M.gguf | Q3_K_M | 5.78GB |
| Rose-2x7B.Q3_K_L.gguf | Q3_K_L | 6.27GB |
| Rose-2x7B.IQ4_XS.gguf | IQ4_XS | 6.5GB |
| Rose-2x7B.Q4_0.gguf | Q4_0 | 6.78GB |
| Rose-2x7B.IQ4_NL.gguf | IQ4_NL | 6.85GB |
| Rose-2x7B.Q4_K_S.gguf | Q4_K_S | 6.84GB |
| Rose-2x7B.Q4_K.gguf | Q4_K | 7.25GB |
| Rose-2x7B.Q4_K_M.gguf | Q4_K_M | 7.25GB |
| Rose-2x7B.Q4_1.gguf | Q4_1 | 7.52GB |
| Rose-2x7B.Q5_0.gguf | Q5_0 | 8.26GB |
| Rose-2x7B.Q5_K_S.gguf | Q5_K_S | 8.26GB |
| Rose-2x7B.Q5_K.gguf | Q5_K | 8.51GB |
| Rose-2x7B.Q5_K_M.gguf | Q5_K_M | 8.51GB |
| Rose-2x7B.Q5_1.gguf | Q5_1 | 9.01GB |
| Rose-2x7B.Q6_K.gguf | Q6_K | 9.84GB |
| Rose-2x7B.Q8_0.gguf | Q8_0 | 12.75GB |
Original model description:
license: apache-2.0 tags:
- moe
- frankenmoe
- merge
- mergekit
- maywell/PiVoT-0.1-Starling-LM-RP
- WizardLM/WizardMath-7B-V1.1 base_model:
- maywell/PiVoT-0.1-Starling-LM-RP
- WizardLM/WizardMath-7B-V1.1
Rose-2x7B
Rose-2x7B is a Mixure of Experts (MoE) made with the following models using Mergekit:
mergekit-moe mergekit_moe.yaml merge --copy-tokenizer --device cuda --low-cpu-memory
🧩 Configuration
base_model: uproai/ros-7b-v1
experts:
- source_model: maywell/PiVoT-0.1-Starling-LM-RP
positive_prompts:
- "storywriting"
- "write"
- "scene"
- "story"
- "character"
- source_model: WizardLM/WizardMath-7B-V1.1
positive_prompts:
- "reason"
- "math"
- "mathematics"
- "solve"
- "count"
tokenizer_source: union
💻 Usage
!pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "uproai/Rose-2x7B"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)
messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
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
- Sep 22, 2024
- Updated
- Sep 22, 2024