Hugging Face's logo

BigCode

Enterprise
non-profit
https://www.bigcode-project.org/
BigCodeProject
bigcode-project
Activity Feed Request to join this org

AI & ML interests

None defined yet.

Recent Activity

sted97  authored a paper 2 days ago
What's the Meaning of Superhuman Performance in Today's NLU?
sted97  authored a paper 2 days ago
Truth or Mirage? Towards End-to-End Factuality Evaluation with LLM-OASIS
sted97  authored a paper 2 days ago
LLMs Lost in Translation: M-ALERT uncovers Cross-Linguistic Safety Gaps
View all activity

bigcode's activity

terryyz 
updated 7 datasets 5 days ago
terryyz 
in bigcode/bigcodebench-evaluator 6 days ago

Is it possible to evaluate a subset?

2
#2 opened 9 days ago by
ChenC
anton-l 
posted an update 6 days ago
view post
Post
1964
Introducing 📐𝐅𝐢𝐧𝐞𝐌𝐚𝐭𝐡: the best public math pre-training dataset with 50B+ tokens!
HuggingFaceTB/finemath

Math remains challenging for LLMs and by training on FineMath we see considerable gains over other math datasets, especially on GSM8K and MATH.

We build the dataset by:
🛠️ carefully extracting math data from Common Crawl;
🔎 iteratively filtering and recalling high quality math pages using a classifier trained on synthetic annotations to identify math reasoning and deduction.

We conducted a series of ablations comparing the performance of Llama-3.2-3B-Base after continued pre-training on FineMath and observe notable gains compared to the baseline model and other public math datasets.

We hope this helps advance the performance of LLMs on math and reasoning! 🚀
We’re also releasing all the ablation models as well as the evaluation code.

HuggingFaceTB/finemath-6763fb8f71b6439b653482c2
nicolaschapados 
authored a paper 7 days ago
gasse 
authored a paper 12 days ago
optimass 
authored 2 papers 12 days ago
gasse 
authored a paper 12 days ago
optimass 
authored a paper 12 days ago
terryyz 
updated 3 datasets 12 days ago
terryyz 
updated a Space 13 days ago
xu3kev 
authored a paper 20 days ago