config.json

{
  "architectures": [
    "RoseX1ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_rose_x1.RoseX1Config",
    "AutoModelForCausalLM": "modeling_rose_x1.RoseX1ForCausalLM"
  },
  "bos_token_id": 2,
  "dtype": "float32",
  "eos_token_id": 3,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 512,
  "initializer_range": 0.02,
  "intermediate_size": 1408,
  "max_position_embeddings": 1024,
  "mlp_bias": false,
  "model_type": "rose_x1",
  "num_attention_heads": 8,
  "num_hidden_layers": 14,
  "num_key_value_heads": 2,
  "pad_token_id": 0,
  "pretraining_tp": 1,
  "refresh_gate_enabled": true,
  "refresh_gate_inject_layers": [
    4,
    9
  ],
  "refresh_gate_kernel_size": 9,
  "rms_norm_eps": 1e-05,
  "rope_parameters": {
    "rope_theta": 10000.0,
    "rope_type": "default"
  },
  "rope_theta": 100000.0,
  "tie_word_embeddings": true,
  "transformers_version": "5.0.0",
  "use_cache": false,
  "use_qk_norm": true,
  "vocab_size": 16384
}