File size: 1,692 Bytes
beeb7a8 91d18f6 beeb7a8 91d18f6 beeb7a8 91d18f6 beeb7a8 91d18f6 beeb7a8 91d18f6 beeb7a8 |
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 |
---
tags:
- merge
- mergekit
- lazymergekit
- microsoft/rho-math-7b-interpreter-v0.1
- WizardLMTeam/WizardMath-7B-V1.1
base_model:
- microsoft/rho-math-7b-interpreter-v0.1
- WizardLMTeam/WizardMath-7B-V1.1
---
# ganeet-V3
ganeet-V3 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [microsoft/rho-math-7b-interpreter-v0.1](https://huggingface.co./microsoft/rho-math-7b-interpreter-v0.1)
* [WizardLMTeam/WizardMath-7B-V1.1](https://huggingface.co./WizardLMTeam/WizardMath-7B-V1.1)
## 🧩 Configuration
```yaml
slices:
- sources:
- model: microsoft/rho-math-7b-interpreter-v0.1
layer_range: [0, 32]
- model: WizardLMTeam/WizardMath-7B-V1.1
layer_range: [0, 32]
merge_method: slerp
base_model: microsoft/rho-math-7b-interpreter-v0.1
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: 0.4
dtype: bfloat16
```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "snigdhachandan/ganeet-V3"
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"])
``` |