{ "architectures": [ "BDH" ], "vocab_size": 256, "n_layer": 8, "n_embd": 256, "n_head": 4, "mlp_internal_dim_multiplier": 128, "dropout": 0.1, "seq_len": 2048, "model_type": "bdh", "size_m": 25, "training": { "steps": 10000, "data": "SlayerLab/polish-dynaword-mix" } }