neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/wav2vec2/hf-internal-testing/tiny-random-Wav2Vec2Model/fe61ce3dd9af75901342.json

{
  "_entry_class": "SingleModelCacheEntry",
  "_model_id": "hf-internal-testing/tiny-random-Wav2Vec2Model",
  "_task": "audio-frame-classification",
  "activation_dropout": 0.1,
  "adapter_attn_dim": null,
  "adapter_kernel_size": 3,
  "adapter_stride": 2,
  "add_adapter": false,
  "apply_spec_augment": true,
  "architectures": [
    "Wav2Vec2Model"
  ],
  "attention_dropout": 0.1,
  "classifier_proj_size": 256,
  "codevector_dim": 256,
  "contrastive_logits_temperature": 0.1,
  "conv_bias": false,
  "conv_dim": [
    32,
    32,
    32
  ],
  "conv_kernel": [
    8,
    8,
    8
  ],
  "conv_stride": [
    4,
    4,
    4
  ],
  "ctc_loss_reduction": "sum",
  "ctc_zero_infinity": false,
  "diversity_loss_weight": 0.1,
  "do_stable_layer_norm": false,
  "feat_extract_activation": "gelu",
  "feat_extract_dropout": 0.0,
  "feat_extract_norm": "group",
  "feat_proj_dropout": 0.0,
  "feat_quantizer_dropout": 0.0,
  "final_dropout": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout": 0.1,
  "hidden_dropout_prob": 0.1,
  "hidden_size": 16,
  "initializer_range": 0.02,
  "intermediate_size": 20,
  "layer_norm_eps": 1e-05,
  "layerdrop": 0.1,
  "mask_feature_length": 10,
  "mask_feature_min_masks": 0,
  "mask_feature_prob": 0.0,
  "mask_time_length": 2,
  "mask_time_min_masks": 2,
  "mask_time_prob": 0.5,
  "model_type": "wav2vec2",
  "neuron": {
    "auto_cast": null,
    "auto_cast_type": null,
    "compiler_type": "neuronx-cc",
    "compiler_version": "2.19.8089.0+8ab9f450",
    "disable_fallback": false,
    "disable_fast_relayout": false,
    "dynamic_batch_size": true,
    "float_dtype": "fp32",
    "inline_weights_to_neff": true,
    "input_names": [
      "input_values"
    ],
    "int_dtype": "int64",
    "model_type": "wav2vec2",
    "optlevel": "2",
    "output_attentions": false,
    "output_hidden_states": false,
    "output_names": [
      "logits"
    ],
    "static_audio_sequence_length": 100000,
    "static_batch_size": 1,
    "task": "audio-frame-classification",
    "tensor_parallel_size": 1
  },
  "num_adapter_layers": 1,
  "num_attention_heads": 2,
  "num_codevector_groups": 2,
  "num_codevectors_per_group": 320,
  "num_conv_pos_embedding_groups": 2,
  "num_conv_pos_embeddings": 16,
  "num_feat_extract_layers": 3,
  "num_hidden_layers": 4,
  "num_negatives": 100,
  "output_hidden_size": 16,
  "proj_codevector_dim": 256,
  "tdnn_dilation": [
    1,
    2
  ],
  "tdnn_dim": [
    32,
    32
  ],
  "tdnn_kernel": [
    5,
    3
  ],
  "use_weighted_layer_sum": false,
  "vocab_size": 32,
  "xvector_output_dim": 32
}