{
  "_name_or_path": "mega-ar-126m-4k",
  "activation": "silu",
  "add_lm_hidden_dense_layer": false,
  "add_token_type_embeddings": false,
  "architectures": [
    "MegaForCausalLM"
  ],
  "attention_activation": "softmax",
  "attention_probs_dropout_prob": 0,
  "bidirectional": false,
  "bos_token_id": 0,
  "chunk_size": -1,
  "classifier_dropout": null,
  "dropout_prob": 0,
  "ema_beta_range": 0.02,
  "ema_delta_alpha_range": 0.2,
  "ema_gamma_omega_range": 1.0,
  "ema_projection_size": 16,
  "eos_token_id": 0,
  "hidden_dropout_prob": 0,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 1536,
  "is_decoder": true,
  "max_positions": 4096,
  "model_type": "mega",
  "nffn_activation_dropout_prob": 0,
  "nffn_hidden_size": 1536,
  "norm_affine": true,
  "normalization_type": "layernorm",
  "normalize_before_ffn": true,
  "normalize_before_mega": true,
  "num_attention_heads": 1,
  "num_hidden_layers": 12,
  "pad_token_id": null,
  "quantization_config": {
    "_load_in_4bit": false,
    "_load_in_8bit": true,
    "bnb_4bit_compute_dtype": "float32",
    "bnb_4bit_quant_storage": "uint8",
    "bnb_4bit_quant_type": "fp4",
    "bnb_4bit_use_double_quant": false,
    "llm_int8_enable_fp32_cpu_offload": false,
    "llm_int8_has_fp16_weight": false,
    "llm_int8_skip_modules": null,
    "llm_int8_threshold": 6.0,
    "load_in_4bit": false,
    "load_in_8bit": true,
    "quant_method": "bitsandbytes"
  },
  "relative_positional_bias": "rotary",
  "shared_representation_size": 192,
  "torch_dtype": "float16",
  "transformers_version": "4.42.4",
  "truncation": null,
  "type_vocab_size": 2,
  "use_cache": true,
  "use_chunking": false,
  "use_feature_dropout": false,
  "use_normalized_ffn": true,
  "vocab_size": 50304
}