GPTJModel class - gptj library

Constructors

blocks → List<GPTJBlock>: final
config → GPTJConfig: final
finalLn → LayerNorm: final
hashCode → int: The hash code for this object.
no setterinherited
lmHead → Linear: final
ropeByDevice → Map<Device, RopeCache>: One RopeCache per unique device that hosts at least one transformer block. Blocks look up their cache by device.
final
runtimeType → Type: A representation of the runtime type of the object.
no setterinherited
training ↔ bool: Whether this module is in training mode. Layers that behave differently between training and inference (e.g. Dropout) read this flag in their call method. Defaults to training mode.
getter/setter pairinherited
wte → Embedding: final

call(Tensor tokens) → Tensor: Forward pass. tokens is a 1D [seqLen] float tensor of token ids. Output is [seqLen, vocab] logits.
eval() → void: Put this module (and any registered submodules) into evaluation mode.
inherited
generate(List<double> prompt, {required int maxNewTokens, double temperature = 1.0, int? topK, Random? rng, bool useCache = true}) → List<double>: Autoregressive sampling — same interface as PythiaModel.generate.
noSuchMethod(Invocation invocation) → dynamic: Invoked when a nonexistent method or property is accessed.
inherited
parameters() → List<Tensor>: Trainable tensors owned by this module (and its submodules).
override
submodules() → List<Module>: Submodules owned by this module. Subclasses that compose other modules should override this so train() / eval() propagate. Default: empty.
override
toString() → String: A string representation of this object.
inherited
train() → void: Put this module (and any registered submodules) into training mode.
inherited
zeroGrad() → void: Zero every parameter's gradient. Safe to call before each backward.
inherited