esm2_t6_8M / README.md
prasadsachin's picture
Upload folder using huggingface_hub
aa89982 verified
metadata
library_name: keras-hub

This is a ESM model uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends. This model is related to a MaskedPLM task.

Model config:

  • name: esm_backbone
  • trainable: True
  • vocabulary_size: 33
  • num_layers: 6
  • num_heads: 20
  • hidden_dim: 320
  • intermediate_dim: 1280
  • dropout: 0.0
  • max_wavelength: 10000
  • use_bias: True
  • activation: gelu
  • layer_norm_eps: 1e-05
  • use_pre_layer_norm: False
  • position_embedding_type: rotary
  • max_sequence_length: 1026
  • pad_token_id: 1

This model card has been generated automatically and should be completed by the model author. See Model Cards documentation for more information.