transformer_lens.model_bridge.supported_architectures.minimax_m2 module

MiniMax-M2 architecture adapter.

class transformer_lens.model_bridge.supported_architectures.minimax_m2.MiniMaxM2ArchitectureAdapter(cfg: Any)

Bases: ArchitectureAdapter

Architecture adapter for MiniMaxM2ForCausalLM models – Qwen3-MoE-like, but with full-width (not per-head) Q/K norm and a sigmoid + e_score_correction_bias router.

__init__(cfg: Any) → None

Initialize the MiniMax-M2 architecture adapter.