model.safetensors -> 200K-step weights (val PPL 2.4106), replaces 60K (2.556) 582fd06 verified EverestAn commited on 1 day ago
upload 200K-step baseline checkpoint (raw PyTorch, bf16, weight-tied) c8fa9d1 verified EverestAn commited on 1 day ago
M2-2B: 30K-step byte-level checkpoint (val PPL 2.73, bf16) 99ff3bb verified EverestAn commited on 6 days ago
M2-2B: 30K-step byte-level checkpoint (val PPL 2.73, bf16) edd793c verified EverestAn commited on 12 days ago
M2-2B: 30K-step byte-level checkpoint (val PPL 2.73, bf16) aca3407 verified EverestAn commited on 12 days ago
M2-2B: 30K-step byte-level checkpoint (val PPL 2.73, bf16) 3a59b32 verified EverestAn commited on 14 days ago
M2-2B: 30K-step byte-level checkpoint (val PPL 2.73, bf16) 8d0fec6 verified EverestAn commited on 14 days ago