config.json

{
  "_name_or_path": "./",
  "architectures": [
    "HybriDNAForCausalLM"
  ],
  "attention_dropout": 0.0,
  "attn_layer_offset": 4,
  "attn_layer_period": 8,
  "auto_map": {
    "AutoConfig": "configuration_hybridna.HybriDNAConfig",
    "AutoModel": "modeling_hybridna.HybriDNAModel",
    "AutoModelForCausalLM": "modeling_hybridna.HybriDNAForCausalLM"
  },
  "bos_token_id": 2,
  "chunk_size": 256,
  "eos_token_id": 1,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 8192,
  "mamba_conv_bias": true,
  "mamba_d_conv": 4,
  "mamba_d_state": 64,
  "mamba_dt_rank": 64,
  "mamba_expand": 2,
  "mamba_proj_bias": false,
  "max_position_embeddings": 131074,
  "model_type": "hybridna",
  "n_groups": 8,
  "num_attention_heads": 32,
  "num_hidden_layers": 16,
  "num_key_value_heads": 8,
  "num_logits_to_keep": 2,
  "pad_token_id": 4,
  "rms_norm_eps": 1e-06,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "time_step_floor": 0.0001,
  "time_step_limit": [
    0.0,
    Infinity
  ],
  "time_step_max": 0.1,
  "time_step_min": 0.001,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.42.4",
  "use_cache": false,
  "use_mamba_kernels": true,
  "vocab_size": 12
}