Sasi Kiran's picture

1 138

Sasi Kiran

sasikiran

·

AI & ML interests

Large language models

Recent Activity

liked a model 1 day ago

tencent/WeDLM-8B-Instruct

liked a model about 1 month ago

nvidia/Cosmos-Reason2-8B

reacted to codelion's post with 🔥 about 1 month ago

Introducing Dhara-70M: A diffusion language model that achieves 3.8x higher throughput than autoregressive models! Key findings from our research on optimal architectures for small language models: → Depth beats width: 32 layers outperforms 12 layers at the same parameter count → Best-in-class factuality: 47.5% on TruthfulQA → 10x training efficiency using WSD (Warmup-Stable-Decay) conversion → Canon layers add only 0.13% parameters but improve reasoning We trained on 1B tokens using the optimal 50-30-20 dataset mix (PDFs + filtered web + educational content), then converted to diffusion with just 100M additional tokens. Blog: https://huggingface.co/blog/codelion/optimal-model-architecture Model: https://huggingface.co/codelion/dhara-70m

View all activity

Organizations

sasikiran 's datasets

None public yet