File size: 1,630 Bytes
8fd31cc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
---
license: cc-by-nc-4.0
tags:
- merge
- mergekit
- lazymergekit
- automerger
base_model:
- Kukedlc/NeuralMiLLaMa-8B-slerp
- KingNish/KingNish-Llama3-8b
---
# NeuralmillamaKingnish-8B
NeuralmillamaKingnish-8B is an automated merge created by [Maxime Labonne](https://huggingface.co/mlabonne) using the following configuration.
* [Kukedlc/NeuralMiLLaMa-8B-slerp](https://huggingface.co/Kukedlc/NeuralMiLLaMa-8B-slerp)
* [KingNish/KingNish-Llama3-8b](https://huggingface.co/KingNish/KingNish-Llama3-8b)
## 🧩 Configuration
```yaml
slices:
- sources:
- model: Kukedlc/NeuralMiLLaMa-8B-slerp
layer_range: [0, 32]
- model: KingNish/KingNish-Llama3-8b
layer_range: [0, 32]
merge_method: slerp
base_model: Kukedlc/NeuralMiLLaMa-8B-slerp
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: 0.5
dtype: bfloat16
random_seed: 0
```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "automerger/NeuralmillamaKingnish-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"])
``` |