config.json

{
  "architectures": [
    "Lfm2ForCausalLM"
  ],
  "block_auto_adjust_ff_dim": false,
  "block_dim": 2048,
  "block_ff_dim": 10752,
  "block_ffn_dim_multiplier": 1.0,
  "block_mlp_init_scale": 1.0,
  "block_multiple_of": 256,
  "block_norm_eps": 1e-05,
  "block_out_init_scale": 1.0,
  "block_use_swiglu": true,
  "block_use_xavier_init": true,
  "bos_token_id": 1,
  "conv_L_cache": 3,
  "conv_bias": false,
  "conv_dim": 2048,
  "conv_dim_out": 2048,
  "conv_use_xavier_init": true,
  "eos_token_id": 7,
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 10752,
  "layer_types": [
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv",
    "full_attention",
    "conv",
    "conv"
  ],
  "max_position_embeddings": 128000,
  "model_type": "lfm2",
  "norm_eps": 1e-05,
  "num_attention_heads": 32,
  "num_heads": 32,
  "head_dim": 64,
  "num_hidden_layers": 30,
  "num_key_value_heads": 8,
  "pad_token_id": 0,
  "rope_theta": 1000000.0,
  "theta": 1000000.0,
  "tie_embedding": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.55.0.dev0",
  "use_cache": true,
  "use_pos_enc": true,
  "addr_l_begin_mha": 20480,
  "addr_l_end_mha": 35328,
  "flm_version": "0.9.24",
  "vocab_size": 65536
}