community-projects

125 threads · Page 2 of 3

Scaling Laws for Emergent Misalignment 13 messages
Truth-First Alignment (Beginner Friendly project) 10 messages
Methods for Reducing Chain-of-Thought Redundancy 16 messages
Looking to collaborate on projects. 11 messages
Reason-Action model 35 messages
NLP
Fast Approximate Sparse Attention 3 messages
Agentic workflow to automate math problem solving 2 messages
Energy consumption and co2 emissions of the models 11 messages
Situational Awareness through the time. 5 messages
An easy intro to GPT-NeoX in Colab 31 messages
RWKV-6 but with complex log space calculations 993 messages
Fast Orthogonal Reparamaterization 8 messages
Designing and training a llama-1 scale model with forecasting 15 messages
Understanding Hidden Computations in Chain-of-Thought Reasoning 14 messages
Decrypting hidden chain of thought 4 messages
Finetune audio i/o onto Llama 3 163 messages
Gradient models: generating data through gradient descent 35 messages
Goldfish Loss paper replication & experiments 4 messages
NLP
arc kaggle thingy ̶2̶0̶2̶4̶ 2̶0̶2̶5̶ 2026 228 messages
Other Modalities
Rhythm Game Map Generation 3 messages
Other Modalities
Support for MLX models in lm-evaluation-harness (PR #1902) 37 messages
NLP
LLM for Graph Structural Learning 4 messages
HuRR/DERP: Human Researcher Respite through Dynamic Embeddings Representation Projection 2 messages
Alignment
Unicorn test 94 messages
Distilling/Training a 100M/1B LLM (maybe from LLaMA-3?) 18 messages
NLP
STRIPS + LLMs friendly AI layer 19 messages
Other Modalities Alignment Interpretability
aimo progress prize 1 155 messages
LlamaTokenizer (without sentencepiece) 2 messages
NLP
Fixing MMLU 38 messages
NLP
t-jepa 101 messages
NLP
DipoleAttention 15 messages
side channel attacks 26 messages
Model Adaptation with Diffusion 27 messages
Advancing Mathematical Problem-Solving in Language Models through Latent Intermediate Reasoning 14 messages
NLP
Dynamic NTK Isn't Correct 203 messages
What is learned, when? 1413 messages
Extending Universal Transformers 62 messages
NLP
A part 2 paper of role playing with LLMs: autism edition 7 messages
NLP Alignment
Training LLM's on Mafia/werewolf deception based- text games 215 messages
NLP
Systematic Investigation of Mu-Transfer 817 messages
Understanding Cross-Modal Compression in Language Models 6 messages
Other Modalities Interpretability
Web & desktop navigation agents and datasets 7 messages
Model size vs Pretraining wrt weak-to-strong generalization 13 messages
Let's end with BPE & pretokenizers 88 messages
Mamba training in axolotl 9 messages
GπT 11 messages
Other Modalities
Large Language Models for Chemistry / Drug Design 37 messages
Harden FFF boundaries during training for large speedup 303 messages
Learning rates and init 278 messages
RadialLayer, an alternative to Fast Feedforward layers 39 messages