{ "architectures": [ "SkipMiddleModel" ], "dim": 768, "ffn_dim_multiplier": 4, "initializer_range": 0.02, "max_seq_len": 1024, "multiple_of": 256, "n_heads": 12, "n_kv_heads": 12, "n_layers": 8, "norm_eps": 1e-05, "rope_theta": 10000, "torch_dtype": "bfloat16", "transformers_version": "4.51.1", "use_scaled_rope": false, "vocab_size": 50257, "zero_init_masks": true }