tensorflow/models · error · ValueError

Unknown `output` value "%s". `output` can be either "logits"

Error message

Unknown `output` value "%s". `output` can be either "logits" or "predictions"

What it means

Error "Unknown `output` value "%s". `output` can be either "logits" or "predictions"" thrown in tensorflow/models.

Source

Thrown at official/projects/teams/teams_pretrainer.py:74

              num_attention_heads=self.hidden_cfg['num_attention_heads'],
              intermediate_size=self.hidden_cfg['intermediate_size'],
              intermediate_activation=self.activation,
              dropout_rate=self.hidden_cfg['dropout_rate'],
              attention_dropout_rate=self.hidden_cfg['attention_dropout_rate'],
              kernel_initializer=tf_utils.clone_initializer(self.initializer),
              name='transformer/layer_%d_rtd' % i))
    self.dense = tf_keras.layers.Dense(
        self.hidden_size,
        activation=self.activation,
        kernel_initializer=tf_utils.clone_initializer(self.initializer),
        name='transform/rtd_dense')
    self.rtd_head = tf_keras.layers.Dense(
        units=1,
        kernel_initializer=tf_utils.clone_initializer(self.initializer),
        name='transform/rtd_head')

    if output not in ('predictions', 'logits'):
      raise ValueError(
          ('Unknown `output` value "%s". `output` can be either "logits" or '
           '"predictions"') % output)
    self._output_type = output

  def call(self, sequence_data, input_mask):
    """Compute inner-products of hidden vectors with sampled element embeddings.

    Args:
      sequence_data: A [batch_size, seq_length, num_hidden] tensor.
      input_mask: A [batch_size, seq_length] binary mask to separate the input
        from the padding.

    Returns:
      A [batch_size, seq_length] tensor.
    """
    attention_mask = layers.SelfAttentionMask()([sequence_data, input_mask])
    data = sequence_data
    for hidden_layer in self.hidden_layers:

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/teams/teams_pretrainer.py:74 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tensorflow/models@e006f5f0d5 (2026-08-24). Data as JSON: /api/errors/af45c70f23ab06d7. Report an issue: GitHub.