config.json

{
  "architectures": [
    "RoseX15ForCausalLM"
  ],
  "model_type": "rose_x1_5",
  "auto_map": {
    "AutoConfig": "configuration_rose_x1_5.RoseX15Config",
    "AutoModelForCausalLM": "modeling_rose_x1_5.RoseX15ForCausalLM"
  },
  "vocab_size": 32770,
  "hidden_size": 640,
  "intermediate_size": 1536,
  "num_hidden_layers": 30,
  "num_attention_heads": 10,
  "num_key_value_heads": 5,
  "head_dim": 64,
  "max_position_embeddings": 2048,
  "hidden_act": "silu",
  "rms_norm_eps": 1e-06,
  "attention_bias": false,
  "mlp_bias": false,
  "attention_dropout": 0.0,
  "tie_word_embeddings": true,
  "rope_theta": 100000.0,
  "rope_scaling": null,
  "use_qk_norm": true,
  "use_xsa": true,
  "refresh_gate_enabled": true,
  "refresh_gate_inject_layers": [
    9,
    19
  ],
  "refresh_gate_kernel_size": 9,
  "mup_enabled": false,
  "mup_base_hidden_size": null,
  "mup_width_mult": null,
  "bos_token_id": 0,
  "eos_token_id": 0,
  "pad_token_id": 1,
  "torch_dtype": "float32",
  "transformers_version": "4.40.0",
  "use_cache": true,
  "pretraining_tp": 1,
  "initializer_range": 0.02
}