NeuralPipe-7B-slerp
NeuralPipe-7B-slerp is a merge of the following models using LazyMergekit:
- RaduGabriel/MUZD
- RaduGabriel/Mistral-Instruct-Ukrainian-SFT
- Radu1999/MisterUkrainianDPO
- CultriX/NeuralTrix-7B-dpo
🧩 Configuration
models:
- model: RaduGabriel/MUZD
parameters:
weight: 0.3
- model: RaduGabriel/Mistral-Instruct-Ukrainian-SFT
parameters:
weight: 0.3
- model: Radu1999/MisterUkrainianDPO
parameters:
weight: 0.1
- model: CultriX/NeuralTrix-7B-dpo
parameters:
weight: 0.3
merge_method: task_arithmetic
base_model: mistralai/Mistral-7B-v0.1
dtype: bfloat16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "RaduGabriel/SirUkrainian"
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.bfloat16,
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"])
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 70.50 |
AI2 Reasoning Challenge (25-Shot) | 67.32 |
HellaSwag (10-Shot) | 85.54 |
MMLU (5-Shot) | 63.14 |
TruthfulQA (0-shot) | 68.74 |
Winogrande (5-shot) | 81.53 |
GSM8k (5-shot) | 56.71 |
- Downloads last month
- 89
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for RaduGabriel/SirUkrainian
Merge model
this model
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard67.320
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard85.540
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard63.140
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard68.740
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard81.530
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard56.710