{ "vocab": 8000, "d_model": 256, "heads": 8, "d_ff": 1536, "enc_layers": 8, "dec_layers": 4, "dropout": 0.1, "max_len": 192, "model_type": "teencode-seq2seq-transformer", "step": 13000, "params": 19935040, "pad_id": 0, "bos_id": 1, "eos_id": 2, "unk_id": 3, "dtype": "float16", "norm": "pre-LayerNorm" }