config.json

{
  "architectures": [
    "CMAForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "modeling_cma.CMAConfig",
    "AutoModelForCausalLM": "modeling_cma.CMAForCausalLM"
  },
  "model_type": "cma",
  "vocab_size": 4096,
  "seq_len": 1024,
  "max_position_embeddings": 1024,
  "n_positions": 1024,
  "n_ctx": 1024,
  "d_model": 216,
  "n_layers": 10,
  "n_heads": 6,
  "n_kv_heads": 2,
  "chunk": 24,
  "cma_heads": 3,
  "expand": 2,
  "cma_identity_prob": 0.9,
  "route_mode": "signed_residual_channel_attention",
  "cma_conditioning": "query_key_value_with_silu_gate",
  "cma_routing_backend": "dense_chunk_softmax",
  "tokenizer_source": "AxiomicLabs/GPT-S-5M",
  "tokenizer_revision": "275b9c3ca78736bf6aeb154c7e2d5f5764fe9035",
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 2,
  "tie_word_embeddings": true,
  "is_decoder": true,
  "is_encoder_decoder": false,
  "use_cache": false,
  "safe_serialization": true,
  "minimum_torch_version": "2.5",
  "torch_dtype": "bfloat16",
  "transformers_version": "5.14.1"
}