- Muon for 2D matrix params, AdamW for 1D (norm/bias/embed) - MuonMix wrapper handles combined step/zero_grad/state_dict - New CLI args: weight_decay, muon_momentum, muon_nesterov, muon_ns_steps, muon_adjust_lr - Removed adamw_beta1/adamw_beta2/adamw_weight_decay - Moved optimizer/strategy params from signature to **kwargs |
||
|---|---|---|
| .. | ||
| demo | ||
| eval | ||
| tools | ||
| docker.sh | ||
| pre_commit.sh | ||