{"record":{"id":"9d36e1f9ee87fa58","repo":"tensorflow/models","slug":"the-input-size-d-is-not-a-multiple-of-the-numbe-9d36e1","errorCode":null,"errorMessage":"The input size (%d) is not a multiple of the number of attention heads (%d)","messagePattern":"The input size \\((.+?)\\) is not a multiple of the number of attention heads \\((.+?)\\)","errorType":"exception","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"official/nlp/modeling/layers/rezero_transformer.py","lineNumber":169,"sourceCode":"          \"shape [batch, sequence, width].\"\n      )\n    batch_size, sequence_length, hidden_size = input_tensor_shape\n\n    if len(input_shape) == 2:\n      mask_tensor_shape = tf.TensorShape(input_shape[1])\n      expected_mask_tensor_shape = tf.TensorShape(\n          [batch_size, sequence_length, sequence_length]\n      )\n      if not expected_mask_tensor_shape.is_compatible_with(mask_tensor_shape):\n        raise ValueError(\n            \"When passing a mask tensor to TransformerLayer, the \"\n            \"mask tensor must be of shape [batch, \"\n            \"sequence_length, sequence_length] (here %s). Got a \"\n            \"mask tensor of shape %s.\"\n            % (expected_mask_tensor_shape, mask_tensor_shape)\n        )\n    if hidden_size % self._num_heads != 0:\n      raise ValueError(\n          \"The input size (%d) is not a multiple of the number of attention \"\n          \"heads (%d)\" % (hidden_size, self._num_heads)\n      )\n    self._attention_head_size = int(hidden_size // self._num_heads)\n    common_kwargs = dict(\n        kernel_regularizer=self._kernel_regularizer,\n        bias_regularizer=self._bias_regularizer,\n        activity_regularizer=self._activity_regularizer,\n        kernel_constraint=self._kernel_constraint,\n        bias_constraint=self._bias_constraint,\n    )\n    attention_kwargs = dict(\n        num_heads=self._num_heads,\n        key_dim=self._attention_head_size,\n        dropout=self._attention_dropout_rate,\n        name=\"self_attention\",\n        kernel_initializer=tf_utils.clone_initializer(self._kernel_initializer),\n        bias_initializer=tf_utils.clone_initializer(self._bias_initializer),","sourceCodeStart":151,"sourceCodeEnd":187,"githubUrl":"https://github.com/tensorflow/models/blob/e006f5f0d534913e49c1f1dae87364039fa607e2/official/nlp/modeling/layers/rezero_transformer.py#L151-L187","documentation":"Error \"The input size (%d) is not a multiple of the number of attention heads (%d)\" thrown in tensorflow/models.","triggerScenarios":"Thrown at official/nlp/modeling/layers/rezero_transformer.py:169 when the library encounters an invalid state.","commonSituations":"See trigger scenarios.","solutions":[],"exampleFix":null,"handlingStrategy":null,"validationCode":null,"typeGuard":null,"tryCatchPattern":null,"preventionTips":[],"tags":[],"backgroundTag":null,"analyzedSha":"e006f5f0d534913e49c1f1dae87364039fa607e2","analyzedAt":"2026-08-24T14:09:15.576Z","schemaVersion":2},"datasetVersion":"2026-08-24T17:17:21.512Z"}