Modelpublic

aspect-ratio-scaling/hc-lr5e-4-llama-1600M-L54-pretrain

sourceHugging Faceupdated 4d agoView on Hugging Face

0likes31downloads

Model Card

hc-lr5e-4-llama-1600M-L54-pretrain

DepthBench final OLMo-core distributed training checkpoint, step 30400. This is not converted to Transformers/safetensors; AutoModel.from_pretrained is not supported. Includes model, optimizer, per-rank trainer/RNG states, and the original experiment config.

—Method: hc
—Nominal total-size tier: 1600M
—Layers: 54; hidden: 1504; intermediate: 4512
—Peak learning rate: 0.0005
—Experiment seed: 42
—Source run: pretrain-hc-1600M-L54-lr5e4-b200-8gpu-paired-20260916

See config.json for all settings and depthbench-endpoint.json for source file hashes. .metadata.json is published last, after all checkpoint shards have been uploaded. Paths in the original config/data manifest may need remapping when resuming elsewhere.