giansha's picture

5 1 2

giansha

giant-S

·

giansha

AI & ML interests

transformer/Time Series Analysis

Recent Activity

upvoted a paper 6 months ago

A Cookbook of Self-Supervised Learning

new activity 11 months ago

perplexity-ai/r1-1776:Using another biased dataset to correct a biased model, bring serious alignment and safety problems.

new activity 11 months ago

perplexity-ai/r1-1776:good job!

View all activity

Organizations

None yet

upvoted a paper 6 months ago

A Cookbook of Self-Supervised Learning

Paper • 2304.12210 • Published Apr 24, 2023 • 6

New activity in perplexity-ai/r1-1776 11 months ago

Using another biased dataset to correct a biased model, bring serious alignment and safety problems.

#110 opened 11 months ago by

good job!

#117 opened 11 months ago by

🚩 Report: Ethical issue(s)

#114 opened 11 months ago by

F*ck Indian rubbish in Perplexity

#102 opened 11 months ago by

还以为1776是啥关键词呢，搞半天是牢美的独立宣言啊

#108 opened 11 months ago by

reacted to fdaudens's post with 🤯 11 months ago

Post

5872

🎯 Perplexity drops their FIRST open-weight model on Hugging Face: A decensored DeepSeek-R1 with full reasoning capabilities. Tested on 1000+ examples for unbiased responses.

Check it out: perplexity-ai/r1-1776
Blog post: https://perplexity.ai/hub/blog/open-sourcing-r1-1776

1 reply

·

replied to fdaudens's post 11 months ago

Using another biased dataset to correct a biased model, remove censorship will undoubtedly bring more serious alignment and safety problems.

liked a Space 11 months ago

The Ultra-Scale Playbook

The ultimate guide to training LLM on large GPU Clusters

liked a Space over 1 year ago

OpenGPT 4o

GPT 4o like bot.