File size: 1,945 Bytes
6f49091 c0b8022 6f49091 |
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 65 66 67 68 69 70 71 |
---
tags:
- merge
- mergekit
- lazymergekit
- timpal0l/Mistral-7B-v0.1-flashback-v2
- abacusai/Slerp-CM-mist-dpo
- EmbeddedLLM/Mistral-7B-Merge-14-v0.2
base_model:
- timpal0l/Mistral-7B-v0.1-flashback-v2
- abacusai/Slerp-CM-mist-dpo
- EmbeddedLLM/Mistral-7B-Merge-14-v0.2
license: apache-2.0
---
# test-dare
test-dare is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [timpal0l/Mistral-7B-v0.1-flashback-v2](https://huggingface.co/timpal0l/Mistral-7B-v0.1-flashback-v2)
* [abacusai/Slerp-CM-mist-dpo](https://huggingface.co/abacusai/Slerp-CM-mist-dpo)
* [EmbeddedLLM/Mistral-7B-Merge-14-v0.2](https://huggingface.co/EmbeddedLLM/Mistral-7B-Merge-14-v0.2)
## 🧩 Configuration
```yaml
models:
- model: mistralai/Mistral-7B-v0.1
# No parameters necessary for base model
- model: timpal0l/Mistral-7B-v0.1-flashback-v2
parameters:
density: 0.53
weight: 0.4
- model: abacusai/Slerp-CM-mist-dpo
parameters:
density: 0.53
weight: 0.3
- model: EmbeddedLLM/Mistral-7B-Merge-14-v0.2
parameters:
density: 0.53
weight: 0.3
merge_method: dare_ties
base_model: mistralai/Mistral-7B-v0.1
parameters:
int8_mask: true
dtype: bfloat16
```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "FredrikBL/test-dare"
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"])
``` |