81 lines
2.2 KiB
Markdown
81 lines
2.2 KiB
Markdown
|
---
|
||
|
|
tags:
|
||
|
|
- merge
|
||
|
|
- mergekit
|
||
|
|
- lazymergekit
|
||
|
|
- Kukedlc/NeuralSirKrishna-7b
|
||
|
|
- Kukedlc/NeuralKybalion-7B-slerp-v3
|
||
|
|
- Kukedlc/SuperMente-7B-v3
|
||
|
|
base_model:
|
||
|
|
- Kukedlc/NeuralSirKrishna-7b
|
||
|
|
- Kukedlc/NeuralKybalion-7B-slerp-v3
|
||
|
|
- Kukedlc/SuperMente-7B-v3
|
||
|
|
license: apache-2.0
|
||
|
|
---
|
||
|
|
|
||
|
|
# SuperMente-7B-v4
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
SuperMente-7B-v4 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
|
||
|
|
* [Kukedlc/NeuralSirKrishna-7b](https://huggingface.co/Kukedlc/NeuralSirKrishna-7b)
|
||
|
|
* [Kukedlc/NeuralKybalion-7B-slerp-v3](https://huggingface.co/Kukedlc/NeuralKybalion-7B-slerp-v3)
|
||
|
|
* [Kukedlc/SuperMente-7B-v3](https://huggingface.co/Kukedlc/SuperMente-7B-v3)
|
||
|
|
|
||
|
|
## 🧩 Configuration
|
||
|
|
|
||
|
|
```yaml
|
||
|
|
models:
|
||
|
|
- model: Kukedlc/NeuralSirKrishna-7b
|
||
|
|
parameters:
|
||
|
|
density: [1, 0.7, 0.1] # density gradient
|
||
|
|
weight: 1.0
|
||
|
|
- model: Kukedlc/NeuralKybalion-7B-slerp-v3
|
||
|
|
parameters:
|
||
|
|
density: 0.5
|
||
|
|
weight: [0, 0.3, 0.7, 1] # weight gradient
|
||
|
|
- model: Kukedlc/SuperMente-7B-v3
|
||
|
|
parameters:
|
||
|
|
density: 0.33
|
||
|
|
weight:
|
||
|
|
- filter: mlp
|
||
|
|
value: 0.5
|
||
|
|
- value: 0
|
||
|
|
merge_method: ties
|
||
|
|
base_model: Kukedlc/NeuralSirKrishna-7b
|
||
|
|
parameters:
|
||
|
|
normalize: true
|
||
|
|
int8_mask: true
|
||
|
|
dtype: bfloat16
|
||
|
|
```
|
||
|
|
|
||
|
|
## 💻 Usage
|
||
|
|
|
||
|
|
```python
|
||
|
|
!pip install -qU transformers accelerate
|
||
|
|
|
||
|
|
from transformers import AutoTokenizer
|
||
|
|
import transformers
|
||
|
|
import torch
|
||
|
|
|
||
|
|
model = "Kukedlc/SuperMente-7B-v4"
|
||
|
|
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"])
|
||
|
|
```
|
||
|
|
|
||
|
|
## Model Family
|
||
|
|
|
||
|
|
|
||
|
|
|