tensorflow/models · error · ValueError

reuse_attention should be between -1 and %d in call to %s.

Error message

reuse_attention should be between -1 and %d in call to %s.

What it means

Error "reuse_attention should be between -1 and %d in call to %s." thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/layers/reuse_attention.py:232

               pe_max_seq_length=512,
               use_bias=True,
               output_shape=None,
               attention_axes=None,
               kernel_initializer="glorot_uniform",
               bias_initializer="zeros",
               kernel_regularizer=None,
               bias_regularizer=None,
               activity_regularizer=None,
               kernel_constraint=None,
               bias_constraint=None,
               **kwargs):
    super().__init__(**kwargs)
    self._num_heads = num_heads
    self._key_dim = key_dim
    self._value_dim = value_dim if value_dim else key_dim
    self._dropout = dropout
    if reuse_attention > self._num_heads or reuse_attention < -1:
      raise ValueError("reuse_attention should be between -1 "
                       "and %d in call to %s." % (self.__class__,
                                                  self._num_heads))
    if reuse_attention == -1:
      reuse_attention = self._num_heads
    self._reuse_heads = reuse_attention
    self._use_relative_pe = use_relative_pe
    self._pe_max_seq_length = pe_max_seq_length
    self._use_bias = use_bias
    self._output_shape = output_shape
    self._kernel_initializer = tf_keras.initializers.get(kernel_initializer)
    self._bias_initializer = tf_keras.initializers.get(bias_initializer)
    self._kernel_regularizer = tf_keras.regularizers.get(kernel_regularizer)
    self._bias_regularizer = tf_keras.regularizers.get(bias_regularizer)
    self._kernel_constraint = tf_keras.constraints.get(kernel_constraint)
    self._bias_constraint = tf_keras.constraints.get(bias_constraint)
    if attention_axes is not None and not isinstance(attention_axes,
                                                     collections.abc.Sized):
      self._attention_axes = (attention_axes,)

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/layers/reuse_attention.py:232 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/0a35319c1f125e69. Report an issue: GitHub.