| { | |
| "vocab_size": 32768, | |
| "n_layers": 18, | |
| "n_heads": 14, | |
| "n_kv_heads": 2, | |
| "d_model": 1792, | |
| "d_ffn": 4864, | |
| "max_seq_len": 2048, | |
| "rope_theta": 1000000.0, | |
| "nope_every": 4, | |
| "rms_eps": 1e-06, | |
| "init_std": 0.02, | |
| "dropout": 0.0, | |
| "tie_embeddings": true, | |
| "qk_norm": false, | |
| "z_loss_coef": 0.0001 | |
| } |