If you're interested in my models, you can vote for them to be evaluated on the Leadarboard.
GutenLaserPi
GutenLaserPi is a merge of the following models using LazyMergekit:
𧩠Configuration
slices:
- sources:
- model: flammenai/flammen15-gutenberg-DPO-v1-7B
layer_range: [0, 32]
- model: Eric111/CatunaLaserPi
layer_range: [0, 32]
merge_method: slerp
base_model: Eric111/CatunaLaserPi
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: 0.5
dtype: bfloat16
π» Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Stark2008/GutenLaserPi"
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"])
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 21.29 |
IFEval (0-Shot) | 42.27 |
BBH (3-Shot) | 32.98 |
MATH Lvl 5 (4-Shot) | 7.18 |
GPQA (0-shot) | 4.92 |
MuSR (0-shot) | 16.99 |
MMLU-PRO (5-shot) | 23.40 |
- Downloads last month
- 19
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 Stark2008/GutenLaserPi
Merge model
this model
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard42.270
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard32.980
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard7.180
- acc_norm on GPQA (0-shot)Open LLM Leaderboard4.920
- acc_norm on MuSR (0-shot)Open LLM Leaderboard16.990
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard23.400