config.json

{
  "_use_flash_attention_2": true,
  "add_linear_bias": false,
  "apply_rotary_positional_embedding_kernel": false,
  "architectures": [
    "StFoundation2DForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.0,
  "bin_alpha": 1.0,
  "bin_num": 100,
  "dropout": 0.0,
  "hidden_act": "swiglu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 128,
  "initializer_range": 0.02,
  "intermediate_size": 320,
  "layer_norm_eps": 1e-05,
  "mask_token_id": -1,
  "max_position_embeddings": 19264,
  "model_type": "stfoundation2D",
  "normalization_type": "LayerNorm",
  "num_attention_heads": 2,
  "num_hidden_layers": 4,
  "pad_token_id": -2,
  "position_embedding_type": "absolute",
  "rotary_embedding_2d": true,
  "rotary_percent": 1.0,
  "self_attn_type": "core",
  "seq_len_interpolation_factor": null,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.38.0.dev0",
  "type_vocab_size": 2,
  "use_cache": true,
  "vocab_size": 16
}