Zinan Tang

Word2Li's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 2 days ago

MathFusion: Enhancing Mathematic Problem-solving of LLM through Instruction Fusion

Paper • 2503.16212 • Published 3 days ago • 17

upvoted a paper 3 days ago

MetaLadder: Ascending Mathematical Solution Quality via Analogical-Problem Reasoning Transfer

Paper • 2503.14891 • Published 4 days ago • 19

liked a dataset about 1 month ago

QingyiSi/Alpaca-CoT

Preview • Updated Sep 14, 2023 • 12.2k • 725

updated a model 4 months ago

Word2Li/LLaMA-3.1-8B-AGNews-SFT

Text Classification • Updated Dec 5, 2024 • 30

liked 2 datasets 5 months ago

AdamCodd/no_robots-alpaca

Viewer • Updated Jun 17, 2024 • 11.8k • 93 • 7

vicgalle/alpaca-gpt4

Viewer • Updated Feb 10, 2024 • 52k • 11.5k • 279

reacted to beomi's post with 🔥 5 months ago

view post

Post

6370

# PyTorch == 2.5.0 Breaks Transformers' SDPAttention!

When you encounter "RuntimeError: cuDNN Frontend error: [cudnn_frontend] Error: No execution plans support the graph."

We can use workaround like this:

torch.backends.cuda.enable_cudnn_sdp(False)

but this slow downs the performance gain from PyTorch 2.5.

Although it is fixed(not "fixed" but default option is turn-off the cuDNN SDPA) at here -- https://github.com/pytorch/pytorch/pull/138587 , but not released yet. (you need to install directly from source)

Fastest way for now : pip install "torch<2.5"

Ref: https://github.com/huggingface/diffusers/issues/9704#issuecomment-2422585273

🔥

11

11

😎

4

4

👍

2

2

🤯

1

1

😔

1

1

+

Reply

liked 2 models 5 months ago

meta-llama/Llama-3.1-8B

Text Generation • Updated Oct 16, 2024 • 1.1M • 1.51k

Qwen/Qwen2.5-7B

Text Generation • Updated Sep 25, 2024 • 361k • 154

liked a dataset 5 months ago

hendrycks/competition_math

Updated Jun 8, 2023 • 141

liked a model 6 months ago

Qwen/Qwen2-7B

Text Generation • Updated Jun 6, 2024 • 58.3k • 154

liked a model 8 months ago

facebook/opt-350m

Text Generation • Updated Sep 15, 2023 • 258k • • 140

liked a dataset 8 months ago

princeton-nlp/CharXiv

Viewer • Updated Sep 27, 2024 • 2.32k • 1.18k • 32

liked a model 8 months ago

kppkkp/OneChart

Feature Extraction • Updated Feb 4 • 420 • 13