Popular repositories Loading

1. quadmix quadmix Public

   QuaDMix: Quality-Diversity Balanced Data Selection for Efficient LLM Pretraining

   Python 1 1

2. LearningToCompare_FSL LearningToCompare_FSL Public

   Forked from floodsung/LearningToCompare_FSL

   LeainingToCompare_FSL for python3.6

   Python

3. nanochat-npu nanochat-npu Public

   Ascend NPU (910B / CANN) port of karpathy/nanochat — adds a mid-training stage, a STEM eval suite, and a fast bf16 generation-eval path; training backend of climbmix (unofficial Nemotron-CLIMB repr…

   Python 1

4. data-mixture-insights data-mixture-insights Public

   LLM pretraining data mixture research insights

   1

5. climbmix climbmix Public

   Unofficial reproduction of Nemotron-CLIMB (CLustering-based Iterative Data Mixture Bootstrapping, NeurIPS 2025, arXiv:2504.13161) on Ascend NPUs — 5 rounds, ≈3,900 NPU-hours; learned mixtures beat …

   Python

6. BETR BETR Public

   Python