Calls the intercepted method with the given arguments and kwargs. Arguments: args_and_kwargs: The positional and keyword arguments passed t

Signature for a loss function. Args: model: The structure with parameters, usually a neural network model. state: Arbitrary state man

Signature for a training step. Args: train_state: The current state. **kwargs: Arguments passed to the train step, usually inputs to

Standardizes a named array across the given axes.

Root-mean-square standardizes a named array across the given axes.

Runs each of the sublayers in sequence. Args: value: The input to the layer. Returns: The output of the final sublayer.

Runs each of the sublayers in sequence. Args: value: The input to the layer. Returns: The output of the final sublayer.

Runs each of the sublayers in sequence. Args: value: The input to the layer. Returns: The output of the final sublayer.

Returns the input unchanged.

Checks the structure of the value, then returns it.

Randomly drops out components of the input.

Runs each of the sublayers in sequence, then adds the original input. Args: value: The input to the block. Returns: The sum of t

Applies the attention mask to the input array. Args: x: The input array to mask. Usually the matrix of query-key dot products. Returns

Runs the attention computation. Args: x: The input to the computation, which will be mapped to queries, keys, and values by the sub

Runs the caching attention computation and update the K/V cache state. When called, ``self.kv_cache_end_index`` should be filled with a scala

Retrieves tokens from the embedding table. Args: token_index: A named array of token indices. Axis names of this array must be disj

Retrieves tokens from the embedding table. Args: out_embeddings: The output embeddings that should be mapped to token logits. Shoul

Signature for a generalized linear operator `NamedArray` initializer. This signature attempts to make explicit all of the dimensions used by an

Runs the einsum operation.

Runs the linear operator.

Adds a learned bias to the value.

Scales its input by the scaling factor.

Activates the random stream in a context.

Deactivates the random stream.

Retrieves and wraps an attribute of the underlying module. Args: name: The attribute name to retrieve. Returns: Either an AutoNm

Fallback to get a missing method from an ``EffectRequest``.

Fallback to get a missing method from an HandledEffectRef.

Retrieves slices from an indexer. `NamedArray` and `NamedArrayView` can be indexed in two different ways, depending on whether the axes you w

Constructs a new variable. Args: label: The unique label for this variable. value: The initial value of the variable. metadata:

Constructs a new state variable. Args: value: The initial value of the variable. label: The unique label for this variable. If not pr

Constructs an uninitialized parameter. Args: initializer: Initializer to use. name: Name to use. value_structure: If provided,

Checks that new subclasses of Layer have wrapped ``__call__`` if needed.

Returns the number of elements in the selection.

Renders this object with treescope, on a single line.

Treescope handler for named arrays.

Treescope handler for named arrays.