JSpergel_-_test_tiny_mixtral_only_router_2-exl2

Quantization made by Richard Erkhov.

Github

Discord

Request more models

test_tiny_mixtral_only_router_2 - EXL2

• Model creator: https://huggingface.co/JSpergel/
• Original model: https://huggingface.co/JSpergel/test_tiny_mixtral_only_router_2/

Available sizes

| Branch | Bits | Description | | ----- | ---- | ------- | ------ | ------ | ------ | ------ | ------------ | | 8_0 | 8.0 | Maximum quality that ExLlamaV2 can produce, near unquantized performance. | | 6_5 | 6.5 | Very similar to 8.0, good tradeoff of size vs performance, recommended. | | 5_0 | 5.0 | Slightly lower quality vs 6.5, but usable on 8GB cards. | | 4_25 | 4.25 | GPTQ equivalent bits per weight, slightly higher quality. | | 3_5 | 3.5 | Lower quality, only use if you have to. |

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/JSpergel_-_test_tiny_mixtral_only_router_2-exl2 test_tiny_mixtral_only_router_2-6_5

With huggingface hub:

pip3 install huggingface-hub

To download a specific branch, use the --revision parameter. For example, to download the 6.5 bpw branch: Linux:

huggingface-cli download JSpergel_-_test_tiny_mixtral_only_router_2-exl2 --revision 6_5 --local-dir test_tiny_mixtral_only_router_2-6_5 --local-dir-use-symlinks False

Windows (which apparently doesn't like _ in folders sometimes?):

huggingface-cli download JSpergel_-_test_tiny_mixtral_only_router_2-exl2 --revision 6_5 --local-dir test_tiny_mixtral_only_router_2-6.5 --local-dir-use-symlinks False

Original model description:

license: apache-2.0 tags:

• moe
• frankenmoe
• merge
• mergekit
• lazymergekit
• openaccess-ai-collective/tiny-mistral base_model:
• openaccess-ai-collective/tiny-mistral
• openaccess-ai-collective/tiny-mistral
• openaccess-ai-collective/tiny-mistral
• openaccess-ai-collective/tiny-mistral

---

test_tiny_mixtral_only_router_2

test_tiny_mixtral_only_router_2 is a Mixture of Experts (MoE) made with the following models using LazyMergekit:

• openaccess-ai-collective/tiny-mistral
• openaccess-ai-collective/tiny-mistral
• openaccess-ai-collective/tiny-mistral
• openaccess-ai-collective/tiny-mistral

🧩 Configuration

base_model: openaccess-ai-collective/tiny-mistral
gate_mode: hidden
dtype: bfloat16
experts:
  - source_model: openaccess-ai-collective/tiny-mistral
    positive_prompts:
      - "math"
    # You can add negative_prompts if needed
  - source_model: openaccess-ai-collective/tiny-mistral

    positive_prompts:
      - "science"
  - source_model: openaccess-ai-collective/tiny-mistral
    positive_prompts:
      - "writing"
    # You can add negative_prompts if needed
  - source_model: openaccess-ai-collective/tiny-mistral
    positive_prompts:
      - "general"

💻 Usage

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "JSpergel/test_tiny_mixtral_only_router_2"

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"])