config.json

{
  "architectures": [
    "FalconH1ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attention_in_multiplier": 1.0,
  "attention_out_multiplier": 0.0375,
  "attn_layer_indices": null,
  "bos_token_id": 1,
  "embedding_multiplier": 5.656854249492381,
  "eos_token_id": 11,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 21504,
  "key_multiplier": 0.011048543456039804,
  "lm_head_multiplier": 0.0078125,
  "mamba_chunk_size": 128,
  "mamba_conv_bias": true,
  "mamba_d_conv": 4,
  "mamba_d_head": 128,
  "mamba_d_ssm": 4096,
  "mamba_d_state": 256,
  "mamba_expand": 2,
  "mamba_n_groups": 2,
  "mamba_n_heads": 32,
  "mamba_norm_before_gate": false,
  "mamba_proj_bias": false,
  "mamba_rms_norm": true,
  "mamba_use_mlp": true,
  "max_position_embeddings": 262144,
  "mlp_bias": false,
  "mlp_expansion_factor": 8,
  "mlp_multipliers": [
    0.1767766952966369,
    0.011160714285714284
  ],
  "model_type": "falcon_h1",
  "num_attention_heads": 20,
  "num_hidden_layers": 72,
  "num_key_value_heads": 4,
  "num_logits_to_keep": 1,
  "pad_token_id": 0,
  "projectors_bias": false,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 100000000000.0,
  "ssm_in_multiplier": 0.25,
  "ssm_multipliers": [
    0.3535533905932738,
    0.25,
    0.1767766952966369,
    0.5,
    0.3535533905932738
  ],
  "ssm_out_multiplier": 0.08838834764831845,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.52.0.dev0",
  "use_cache": true,
  "vocab_size": 261120
}