config.json

{
  "model_type": "sn38-nanochrono",
  "mix_gate_dim": 24,
  "bos_token_id": 32759,
  "aux_gate_dim": 12,
  "eos_token_id": 32763,
  "architectures": [
    "NanochronoForCausalLM"
  ],
  "vocab_size": 32768,
  "num_attention_heads": 14,
  "layer_types": [
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention"
  ],
  "attention_multiplier": 1.2,
  "rope_theta": 100000.0,
  "tie_word_embeddings": false,
  "tap_layer": 14,
  "intermediate_size": 7168,
  "pad_token_id": 32763,
  "final_logit_softcapping": 15.0,
  "sliding_window": 512,
  "num_key_value_heads": 14,
  "num_hidden_layers": 28,
  "aux_layers": [
    1,
    3,
    5,
    7,
    9,
    11,
    13,
    15,
    17,
    19,
    21,
    23,
    25,
    27
  ],
  "max_position_embeddings": 2048,
  "use_cache": true,
  "hidden_size": 1792,
  "torch_dtype": "bfloat16"
}