config.json

{
  "architectures": [
    "WordLatentTransformerForCausalLM"
  ],
  "bos_token_id": 2,
  "bytes_decoder": {
    "_name_or_path": "sign/utf8-lm-tiny",
    "architectures": [
      "LlamaForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 2,
    "dtype": "float32",
    "eos_token_id": 3,
    "head_dim": 64,
    "hidden_act": "silu",
    "hidden_size": 256,
    "initializer_range": 0.02,
    "intermediate_size": 640,
    "max_position_embeddings": 2048,
    "mlp_bias": false,
    "model_type": "llama",
    "num_attention_heads": 4,
    "num_hidden_layers": 4,
    "num_key_value_heads": 4,
    "pad_token_id": 0,
    "pretraining_tp": 1,
    "rms_norm_eps": 1e-05,
    "rope_scaling": null,
    "rope_theta": 10000.0,
    "use_cache": true,
    "vocab_size": 256
  },
  "bytes_encoder": {
    "_name_or_path": "prajjwal1/bert-tiny",
    "attention_probs_dropout_prob": 0.1,
    "classifier_dropout": null,
    "hidden_act": "gelu",
    "hidden_dropout_prob": 0.1,
    "hidden_size": 128,
    "initializer_range": 0.02,
    "intermediate_size": 512,
    "layer_norm_eps": 1e-12,
    "max_position_embeddings": 512,
    "model_type": "bert",
    "num_attention_heads": 2,
    "num_hidden_layers": 2,
    "position_embedding_type": "absolute",
    "type_vocab_size": 2,
    "use_cache": true,
    "vocab_size": 30522
  },
  "dtype": "bfloat16",
  "eos_token_id": 3,
  "image_encoder": {
    "hidden_size": 0,
    "model_type": "noop_model"
  },
  "is_decoder": true,
  "latent_transformer": {
    "_name_or_path": "sbintuitions/tiny-lm",
    "architectures": [
      "LlamaForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "dtype": "float32",
    "head_dim": 64,
    "hidden_act": "silu",
    "hidden_size": 256,
    "initializer_range": 0.02,
    "intermediate_size": 640,
    "max_position_embeddings": 2048,
    "mlp_bias": false,
    "model_type": "llama",
    "num_attention_heads": 4,
    "num_hidden_layers": 4,
    "num_key_value_heads": 4,
    "pretraining_tp": 1,
    "rms_norm_eps": 1e-05,
    "rope_scaling": null,
    "rope_theta": 10000.0,
    "use_cache": true,
    "vocab_size": 51200
  },
  "modality_dropout": 0.0,
  "model_type": "welt",
  "num_tokens": 256,
  "pad_token_id": 0,
  "tokenizer_class": "UTF8Tokenizer",
  "transformers_version": "4.57.3",
  "trust_remote_code": false
}