{ "adaptive": true, "attn_type": 0, "clamp_len": 1000, "cutoffs": [ 20000, 40000, 200000 ], "d_embed": 1024, "d_head": 64, "d_inner": 4096, "d_model": 1024, "div_val": 4, "dropatt": 0.0, "dropout": 0.1, "ext_len": 0, "init": "normal", "init_range": 0.01, "init_std": 0.02, "mem_len": 1600, "n_head": 16, "n_layer": 18, "n_token": 267735, "pre_lnorm": false, "proj_init_std": 0.01, "same_length": true, "sample_softmax": -1, "tgt_len": 128, "tie_projs": [ false, true, true, true ], "tie_weight": true, "untie_r": true }