Lamar's picture

2

Lamar

lamaraguilar

·

Aguilar

AI & ML interests

chatgpt

Recent Activity

reacted to anton-l's post with 🚀 6 days ago

Introducing 📐𝐅𝐢𝐧𝐞𝐌𝐚𝐭𝐡: the best public math pre-training dataset with 50B+ tokens! https://huggingface.co./datasets/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. https://huggingface.co./collections/HuggingFaceTB/finemath-6763fb8f71b6439b653482c2

reacted to DualityAI-RebekahBogdanoff's post with 🚀 6 days ago

🚀 Unlock the Power of Synthetic Data for AI Training! 🚀 https://huggingface.co./spaces/DualityAI-RebekahBogdanoff/Synthetic_Data_Object_Detection We’ve uploaded our Cheerios Detector model- a YOLOv8 model trained using synthetic data to recognize cereal boxes in indoor environments. But we know you can make it even more robust! 💡 See the model in action in our “Cheerios detector” space, and then take it a step further by using FalconEditor to create custom synthetic data. https://falcon.duality.ai/secure/documentation?learnWelcome=true&sidebarMode=learn With FalconEditor, you can generate complex, targeted data that will enhance your model's performance and make it even more accurate, adaptable, and robust! 🧠✨ Don’t just use the data — improve it! Create unique scenarios, train for rare edge cases, and employ tailored conditions that push the boundaries of AI training.

reacted to etemiz's post with 👀 6 days ago

As more synthetic datasets are made, we move slowly away from human alignment.

View all activity

Organizations

None yet

models

None public yet

datasets

None public yet