metadata
tags:
- merge
- mergekit
- lazymergekit
- WizardLMTeam/WizardMath-7B-V1.1
- microsoft/rho-math-7b-interpreter-v0.1
- meta-math/MetaMath-Mistral-7B
base_model:
- WizardLMTeam/WizardMath-7B-V1.1
- microsoft/rho-math-7b-interpreter-v0.1
- meta-math/MetaMath-Mistral-7B
ganeet-V4
ganeet-V4 is a merge of the following models using LazyMergekit:
- WizardLMTeam/WizardMath-7B-V1.1
- microsoft/rho-math-7b-interpreter-v0.1
- meta-math/MetaMath-Mistral-7B
🧩 Configuration
models:
- model: WizardLMTeam/WizardMath-7B-V1.1
parameters:
density: 0.5 # fraction of weights in differences from the base model to retain
weight: # weight gradient
- filter: mlp
value: 0.5
- value: 0
- model: upaya07/Arithmo2-Mistral-7B
- model: microsoft/rho-math-7b-interpreter-v0.1
parameters:
density: 0.5
weight: 0.5
- model: meta-math/MetaMath-Mistral-7B
parameters:
density: 0.5
weight: 0.5
merge_method: ties
base_model: upaya07/Arithmo2-Mistral-7B
parameters:
normalize: true
int8_mask: true
dtype: float16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "snigdhachandan/ganeet-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"])