README.md

Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


TurdusBeagle-7B - GGUF
- Model creator: https://huggingface.co/leveldevai/
- Original model: https://huggingface.co/leveldevai/TurdusBeagle-7B/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [TurdusBeagle-7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q2_K.gguf) | Q2_K | 2.53GB |
| [TurdusBeagle-7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.IQ3_XS.gguf) | IQ3_XS | 2.81GB |
| [TurdusBeagle-7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.IQ3_S.gguf) | IQ3_S | 2.96GB |
| [TurdusBeagle-7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q3_K_S.gguf) | Q3_K_S | 2.95GB |
| [TurdusBeagle-7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.IQ3_M.gguf) | IQ3_M | 3.06GB |
| [TurdusBeagle-7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q3_K.gguf) | Q3_K | 3.28GB |
| [TurdusBeagle-7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q3_K_M.gguf) | Q3_K_M | 3.28GB |
| [TurdusBeagle-7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q3_K_L.gguf) | Q3_K_L | 3.56GB |
| [TurdusBeagle-7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.IQ4_XS.gguf) | IQ4_XS | 3.67GB |
| [TurdusBeagle-7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q4_0.gguf) | Q4_0 | 3.83GB |
| [TurdusBeagle-7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.IQ4_NL.gguf) | IQ4_NL | 3.87GB |
| [TurdusBeagle-7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q4_K_S.gguf) | Q4_K_S | 3.86GB |
| [TurdusBeagle-7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q4_K.gguf) | Q4_K | 4.07GB |
| [TurdusBeagle-7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q4_K_M.gguf) | Q4_K_M | 4.07GB |
| [TurdusBeagle-7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q4_1.gguf) | Q4_1 | 4.24GB |
| [TurdusBeagle-7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q5_0.gguf) | Q5_0 | 4.65GB |
| [TurdusBeagle-7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q5_K_S.gguf) | Q5_K_S | 4.65GB |
| [TurdusBeagle-7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q5_K.gguf) | Q5_K | 4.78GB |
| [TurdusBeagle-7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q5_K_M.gguf) | Q5_K_M | 4.78GB |
| [TurdusBeagle-7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q5_1.gguf) | Q5_1 | 5.07GB |
| [TurdusBeagle-7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q6_K.gguf) | Q6_K | 5.53GB |
| [TurdusBeagle-7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/leveldevai_-_TurdusBeagle-7B-gguf/blob/main/TurdusBeagle-7B.Q8_0.gguf) | Q8_0 | 7.17GB |




Original model description:
---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- udkai/Turdus
- mlabonne/NeuralBeagle14-7B
---

# TurdusBeagle-7B

TurdusBeagle-7B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [udkai/Turdus](https://huggingface.co/udkai/Turdus)
* [mlabonne/NeuralBeagle14-7B](https://huggingface.co/mlabonne/NeuralBeagle14-7B)


## 🧩 Configuration

```yaml
slices:
    - sources:
      - model: udkai/Turdus
        layer_range: [0, 32]
      - model: mlabonne/NeuralBeagle14-7B
        layer_range: [0, 32]
merge_method: slerp
base_model: mlabonne/NeuralBeagle14-7B
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

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "leveldevai/TurdusBeagle-7B"
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"])
```