Marc Kovka's picture

Marc Kovka

GPT007

·

AI & ML interests

None yet

Recent Activity

liked a model 19 days ago

gokaygokay/Flux-Watercolor-Strokes-LoRA

reacted to lewtun's post with 🔥 19 days ago

We outperform Llama 70B with Llama 3B on hard math by scaling test-time compute 🔥 How? By combining step-wise reward models with tree search algorithms :) We show that smol models can match or exceed the performance of their much larger siblings when given enough "time to think" We're open sourcing the full recipe and sharing a detailed blog post. In our blog post we cover: 📈 Compute-optimal scaling: How we implemented DeepMind's recipe to boost the mathematical capabilities of open models at test-time. 🎄 Diverse Verifier Tree Search (DVTS): An unpublished extension we developed to the verifier-guided tree search technique. This simple yet effective method improves diversity and delivers better performance, particularly at large test-time compute budgets. 🧭 Search and Learn: A lightweight toolkit for implementing search strategies with LLMs and built for speed with vLLM Here's the links: - Blog post: https://huggingface.co/spaces/HuggingFaceH4/blogpost-scaling-test-time-compute - Code: https://github.com/huggingface/search-and-learn Enjoy!

View all activity

Organizations

None yet

GPT007's activity

upvoted 2 articles 5 months ago

Article

Google releases Gemma 2 2B, ShieldGemma and Gemma Scope

Jul 31, 2024

• 59

Article

XetHub is joining Hugging Face!

Aug 8, 2024

• 81

upvoted 4 collections 5 months ago

ShieldGemma Release

A series of safety classifiers, trained on top of Gemma 2, for developers to filter inputs and outputs of their applications. • 3 items • Updated 24 days ago • 11

Gemma Scope Release

A comprehensive, open suite of sparse autoencoders for Gemma 2 2B and 9B. • 10 items • Updated 24 days ago • 15

Gemma 2 2B Release

The 2.6B parameter version of Gemma 2. • 6 items • Updated 24 days ago • 78

Playground v2.5

2 items • Updated Feb 27, 2024 • 23

upvoted a paper 5 months ago

Qwen2-Audio Technical Report

Paper • 2407.10759 • Published Jul 15, 2024 • 55

upvoted a collection 5 months ago

Llama 3.1

This collection hosts the transformers and original repos of the Llama 3.1, Llama Guard 3 and Prompt Guard models • 11 items • Updated about 1 month ago • 638

upvoted a paper 6 months ago

Compact Language Models via Pruning and Knowledge Distillation

Paper • 2407.14679 • Published Jul 19, 2024 • 39

upvoted a collection 6 months ago

DynMoE Family

DynMoE model checkpoints and paper on huggingface • 4 items • Updated Aug 19, 2024 • 4

upvoted a paper 6 months ago

Scaling Diffusion Transformers to 16 Billion Parameters

Paper • 2407.11633 • Published Jul 16, 2024 • 25

upvoted a collection 6 months ago

DCLM

DCLM Models + Datasets • 6 items • Updated Oct 4, 2024 • 25

upvoted a paper 6 months ago

Training language models to follow instructions with human feedback

Paper • 2203.02155 • Published Mar 4, 2022 • 16

upvoted a collection 6 months ago

Qwen2

Qwen2 language models, including pretrained and instruction-tuned models of 5 sizes, including 0.5B, 1.5B, 7B, 57B-A14B, and 72B. • 39 items • Updated Nov 28, 2024 • 353

upvoted a paper 6 months ago

Video Diffusion Alignment via Reward Gradients

Paper • 2407.08737 • Published Jul 11, 2024 • 48

upvoted an article 6 months ago

Article

Introducing Ghost 8B Beta: A Game-Changing Language Model

By

•

Jul 17, 2024

• 7

upvoted a paper 6 months ago

Self-Rewarding Language Models

Paper • 2401.10020 • Published Jan 18, 2024 • 145

upvoted an article 6 months ago

Article

Train a Llama model from scratch

By

•

Jul 29, 2024

• 48

upvoted 2 papers 6 months ago

Vision language models are blind

Paper • 2407.06581 • Published Jul 9, 2024 • 83

Graph-Based Captioning: Enhancing Visual Descriptions by Interconnecting Region Captions

Paper • 2407.06723 • Published Jul 9, 2024 • 11