config.json

{
  "absolute_position_embedding": false,
  "architectures": [
    "MotifForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_motif.MotifConfig",
    "AutoModelForCausalLM": "modeling_motif.MotifForCausalLM"
  },
  "bos_token_id": 219396,
  "decoder_split_layers": [],
  "dim_model_base": 2048,
  "dim_model_base_attn": 128,
  "dim_model_base_init": 2048,
  "dim_model_base_lmh": 1,
  "dim_model_base_logits": 2048,
  "dim_model_base_lr": 256,
  "encoder_split_layers": [],
  "eos_token_id": 219395,
  "hidden_act": "poly_norm",
  "hidden_size": 2048,
  "initializer_range": 2e-05,
  "intermediate_size": 8192,
  "loss_reduction": "mean",
  "max_position_embeddings": 16384,
  "max_window_layers": 28,
  "model_type": "Motif",
  "moe_intermediate_size": null,
  "moe_layer": false,
  "muP": false,
  "multi_token_heads": null,
  "n_group": null,
  "n_routed_experts": null,
  "norm_topk_prob": null,
  "num_attention_heads": 16,
  "num_hidden_layers": 32,
  "num_key_value_heads": 16,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 500000.0,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "topk_group": null,
  "topk_method": null,
  "torch_dtype": "float32",
  "transformers_version": "4.46.3",
  "use_bias": false,
  "use_cache": true,
  "use_qk_norm": false,
  "use_sliding_window": false,
  "vocab_size": 219520,
  "wesar_weights": false
}