{"record":{"id":"da4f0e34f17ad4f1","repo":"tensorflow/models","slug":"block-sparse-attention-does-not-support-multi-quer","errorCode":null,"errorMessage":"Block sparse attention does not support Multi-query attention. Specify only one of them.","messagePattern":"Block sparse attention does not support Multi-query attention\\. Specify only one of them\\.","errorType":"exception","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"official/nlp/modeling/layers/rezero_transformer.py","lineNumber":132,"sourceCode":"    self._bias_regularizer = tf_keras.regularizers.get(bias_regularizer)\n    self._kernel_constraint = tf_keras.constraints.get(kernel_constraint)\n    self._bias_constraint = tf_keras.constraints.get(bias_constraint)\n    self._use_layer_norm = use_layer_norm\n    self._share_rezero = share_rezero\n    self._num_kv_heads = num_kv_heads\n    self._src_block_size = src_block_size\n    self._tgt_block_size = tgt_block_size\n    self._linformer_dim = linformer_dim\n    self._linformer_shared_kv_projection = linformer_shared_kv_projection\n    self._use_sigmoid_attn = use_sigmoid_attn\n    self._sigmoid_attn_bias = sigmoid_attn_bias\n    if self._linformer_dim is not None or self._use_sigmoid_attn:\n      raise ValueError(\n          \"Linformer and Sigmoid attention are not supported in ReZero\"\n          \" Transformer.\"\n      )\n    if self._num_kv_heads is not None and self._src_block_size is not None:\n      raise ValueError(\n          \"Block sparse attention does not support Multi-query attention.\"\n          \" Specify only one of them.\"\n      )\n\n  def build(self, input_shape):\n    if isinstance(input_shape, tf.TensorShape):\n      input_tensor_shape = input_shape\n    elif isinstance(input_shape, (list, tuple)):\n      input_tensor_shape = tf.TensorShape(input_shape[0])\n    else:\n      raise ValueError(\n          \"The type of input shape argument is not supported, got: %s\"\n          % type(input_shape)\n      )\n\n    if len(input_tensor_shape.as_list()) != 3:\n      raise ValueError(\n          \"TransformerLayer expects a three-dimensional input of \"","sourceCodeStart":114,"sourceCodeEnd":150,"githubUrl":"https://github.com/tensorflow/models/blob/e006f5f0d534913e49c1f1dae87364039fa607e2/official/nlp/modeling/layers/rezero_transformer.py#L114-L150","documentation":"Error \"Block sparse attention does not support Multi-query attention. Specify only one of them.\" thrown in tensorflow/models.","triggerScenarios":"Thrown at official/nlp/modeling/layers/rezero_transformer.py:132 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"}