Class DeepJOrigin
java.lang.Object
io.github.kirstenali.deepj.nn.transformer.DecoderOnlyModel
io.github.kirstenali.deepj.nn.transformer.origin.DeepJOrigin
- All Implemented Interfaces:
CausalLM,Persistable,Trainable
-
Field Summary
Fields inherited from class io.github.kirstenali.deepj.nn.transformer.DecoderOnlyModel
lmHead, normF, stack, tokEmb -
Constructor Summary
ConstructorsConstructorDescriptionDeepJOrigin(DeepJOriginConfig cfg, long seed) DeepJOrigin(DeepJOriginConfig cfg, long seed, ParameterFactory parameters) DeepJOrigin(DeepJOriginConfig cfg, long seed, DType dtype) -
Method Summary
Methods inherited from class io.github.kirstenali.deepj.nn.transformer.DecoderOnlyModel
applyInitScale, forward, forwardBatch, load, metadata, openGeneration, openGeneration, parameters, promptFormat, save, setMetadata, setPromptFormat, supportsBatchedForward, supportsRollingGeneration
-
Constructor Details
-
DeepJOrigin
-
DeepJOrigin
-
DeepJOrigin
-
-
Method Details
-
embed
- Overrides:
embedin classDecoderOnlyModel
-
embed
- Overrides:
embedin classDecoderOnlyModel
-
embedBatch
- Overrides:
embedBatchin classDecoderOnlyModel
-
embeddingParameters
- Overrides:
embeddingParametersin classDecoderOnlyModel
-
gradClipNorm
public float gradClipNorm() -
supportsCachedGeneration
protected boolean supportsCachedGeneration()- Overrides:
supportsCachedGenerationin classDecoderOnlyModel
-
config
- Specified by:
configin classDecoderOnlyModel
-