config.json

{
   "_name_or_path": "checkpoint-pretrain",
  "architectures": [
    "PlantbimoeForMaskedLM"
  ],
  "auto_map": {
    "AutoConfig": "configuration_plantbimoe.PlantbimoeConfig",
    "AutoModel": "modeling_plantbimoe.Plantbimoe",
    "AutoModelForMaskedLM": "modeling_plantbimoe.PlantbimoeForMaskedLM",
    "AutoModelForSequenceClassification": "modeling_plantbimoe.PlantbimoeForSequenceClassification"
  },
  "bidirectional": true,
  "bidirectional_strategy": "add",
  "bidirectional_weight_tie": true,
  "d_model": 512,
  "fused_add_norm": true,
  "initializer_cfg": {
    "initializer_range": 0.02,
    "n_residuals_per_layer": 1,
    "rescale_prenorm_residual": true
  },
  "intermediate_size": 1408,
  "model_type": "plantbimoe",
  "n_layer": 16,
  "norm_epsilon": 1e-05,
  "num_experts": 4,
  "num_experts_per_tok": 1,
  "pad_token_id": 4,
  "pad_vocab_size_multiple": 6,
  "residual_in_fp32": false,
  "rms_norm": true,
  "ssm_cfg": {
    "bias": false,
    "conv_bias": true,
    "d_conv": 4,
    "d_state": 16,
    "dt_init": "random",
    "dt_init_floor": 0.0001,
    "dt_max": 0.1,
    "dt_min": 0.001,
    "dt_rank": "auto",
    "dt_scale": 1.0,
    "expand": 2,
    "use_fast_path": true
  },
  "torch_dtype": "float32",
  "transformers_version": "4.38.1",
  "vocab_size": 12
}