ncp arm — Hierarchical Concept-Diffusion synth run

arm: ncp
params: 50,500,737
steps: 12000
seq_len: 256, batch: 32
final val LM loss: 6.1264
data: PleIAs/SYNTH (en subset: query+reasoning+answer, streaming, gpt2 tokenizer)
optimizer: AdamW lr 0.0001, wd 0.1, cosine, warmup 200

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support