neuron/config.json

{
  "_attn_implementation_autoset": true,
  "_name_or_path": "/tmp/tmpdemcaf9_",
  "architectures": [
    "MPNetForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.1,
  "bos_token_id": 0,
  "eos_token_id": 2,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 514,
  "model_type": "mpnet",
  "neuron": {
    "auto_cast": null,
    "auto_cast_type": null,
    "compiler_type": "neuronx-cc",
    "compiler_version": "2.15.143.0+e39249ad",
    "disable_fallback": false,
    "disable_fast_relayout": false,
    "dynamic_batch_size": false,
    "inline_weights_to_neff": true,
    "input_names": [
      "input_ids",
      "attention_mask"
    ],
    "model_type": "mpnet",
    "optlevel": "2",
    "output_attentions": false,
    "output_hidden_states": false,
    "output_names": [
      "last_hidden_state",
      "pooler_output"
    ],
    "static_batch_size": 1,
    "static_sequence_length": 128,
    "tensor_parallel_size": 1
  },
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "pad_token_id": 1,
  "relative_attention_num_buckets": 32,
  "task": "feature-extraction",
  "torchscript": true,
  "transformers_version": "4.48.3",
  "vocab_size": 30527
}