pulsatrix 1.1.0
Loading...
Searching...
No Matches
pulsatrix::FeedForwardModule Class Reference

fc2(f(fc1(x))) over the last dimension. Shape (..., d_model) -> (..., d_model). More...

#include <feed_forward_module.hpp>

Inheritance diagram for pulsatrix::FeedForwardModule:
Collaboration diagram for pulsatrix::FeedForwardModule:

Public Member Functions

 FeedForwardModule (int64_t d_model, int64_t d_ff, DeviceBackend *backend, ElementwiseOp activation=ElementwiseOp::Gelu, bool bias=true)
 
Tensor backward (const Tensor &grad_output) override
 Computes the gradient w.r.t. this module's input, given the gradient w.r.t. its output. Must be called after forward() – uses state cached from that call.
 
Tensor propagate_relevance (const Tensor &relevance_out, const LRPRuleConfig &config) override
 Computes this module's contribution to LRP relevance propagation.
 
OpType op_type () const override
 A chain of other modules.
 
std::vector< NamedParamRef > named_parameters () override
 This module's trainable parameters, each with its hierarchical name – the one place a module declares its parameters (roadmap FND-1).
 
std::vector< NamedModuleRef > named_children () override
 This module's direct sub-modules, each under the prefix its parameters carry in named_parameters(), in the same order (TRN-7: how ApplyLora() finds a model's Linear layers). Default: none. NamedModules() walks the whole tree.
 
std::vector< NamedBufferRef > named_buffers () override
 This module's buffers: state a checkpoint must save that no optimizer updates.
 
void set_training (bool training) override
 Sets this module's training/eval mode. Defaults to training (matches every mainstream framework's Module default).
 
std::optional< DeviceType > compute_device () const override
 The device this module computes on, so forward() can reject an input on another device before any kernel sees it (roadmap FND-8).
 
LinearModule & fc1 ()
 
ActivationModule & activation ()
 
LinearModule & fc2 ()
 
void release_activations () override
 Frees the tensors forward() kept for backward() and the relevance passes; containers release their children too. Inference that only needs the output calls it after each forward, so a deep model holds one layer's activations at a time (a 1024-token ESM-2 650M pass otherwise keeps over 10 GB of them).
 
- Public Member Functions inherited from pulsatrix::Module
virtual ~Module ()=default
 
Tensor forward (const Tensor &input)
 Runs this module's forward computation.
 
std::pair< Tensor, NodeId > forward_traced (const Tensor &input, NodeId input_node, ComputationGraph &graph, Autograd &autograd)
 Runs forward() while also registering a ComputationGraph node (tagged with this module's op_type(), parented to input_node) and wiring an Autograd backward function that reuses this module's own backward() – the opt-in traced/explainable path, per Phase 2 Mission 0.
 
virtual bool supports_lrp_rule (LRPRule rule) const
 Whether propagate_relevance() implements rule (no silent fallback: callers such as ExplainerContext::relevance_pass() throw rather than run a module on a rule it does not implement).
 
virtual std::vector< ParamRef > parameters ()
 This module's trainable parameters and their gradients, for an optimizer to update uniformly across module types.
 
void set_requires_grad (bool requires_grad, const std::string &prefix="")
 Freezes (false) or unfreezes (true) parameters by name (roadmap FND-2).
 
bool is_training () const
 Whether this module is currently in training mode.
 

Protected Member Functions

Tensor forward_impl (const Tensor &input) override
 

Detailed Description

fc2(f(fc1(x))) over the last dimension. Shape (..., d_model) -> (..., d_model).

Note
Parameters are named fc1.* and fc2.*. LRP: each Linear's own rule (the config's), relevance passed through the activation unchanged.

Constructor & Destructor Documentation

◆ FeedForwardModule()

pulsatrix::FeedForwardModule::FeedForwardModule ( int64_t  d_model,
int64_t  d_ff,
DeviceBackend *  backend,
ElementwiseOp  activation = ElementwiseOp::Gelu,
bool  bias = true 
)
Parameters
activationElementwiseOp::Gelu for BERT and ESM; any ActivationModule accepts.
biasBiases on both projections.
Exceptions
std::invalid_argumentif a width isn't positive, or for an op ActivationModule refuses.

Member Function Documentation

◆ activation()

ActivationModule & pulsatrix::FeedForwardModule::activation ( )
inline

◆ backward()

Tensor pulsatrix::FeedForwardModule::backward ( const Tensor &  grad_output)
overridevirtual

Computes the gradient w.r.t. this module's input, given the gradient w.r.t. its output. Must be called after forward() – uses state cached from that call.

Parameters
grad_outputGradient w.r.t. this module's output.
Returns
Gradient w.r.t. this module's input.
Note
Promoted to the base class in Phase 2 Mission 0 – every existing subclass (LinearModule/ReluModule/Conv2DModule) already implemented this exact signature independently; making it virtual lets graph-wiring code (Module::forward_traced) call it polymorphically through a Module* without knowing the concrete subclass, the same way propagate_relevance already works.

Implements pulsatrix::Module.

◆ compute_device()

std::optional< DeviceType > pulsatrix::FeedForwardModule::compute_device ( ) const
inlineoverridevirtual

The device this module computes on, so forward() can reject an input on another device before any kernel sees it (roadmap FND-8).

Returns
std::nullopt (the default) skips the check: a container whose layers check their own inputs, or a user module written before this existed. Every in-tree layer returns its device; EmbeddingModule doesn't, because it reads its indices through their own backend and so accepts them from any device.

Reimplemented from pulsatrix::Module.

◆ fc1()

LinearModule & pulsatrix::FeedForwardModule::fc1 ( )
inline

◆ fc2()

LinearModule & pulsatrix::FeedForwardModule::fc2 ( )
inline

◆ forward_impl()

Tensor pulsatrix::FeedForwardModule::forward_impl ( const Tensor &  input)
overrideprotectedvirtual
Exceptions
std::invalid_argumentunless the input is rank >= 2 with last dimension d_model.

Implements pulsatrix::Module.

◆ named_buffers()

std::vector< NamedBufferRef > pulsatrix::FeedForwardModule::named_buffers ( )
overridevirtual

This module's buffers: state a checkpoint must save that no optimizer updates.

Returns
{name, tensor} entries, named like named_parameters(), in a fixed order. Default: empty. BatchNormModule reports its running statistics; containers prefix their layers' buffers the way they prefix parameters.

Reimplemented from pulsatrix::Module.

◆ named_children()

std::vector< NamedModuleRef > pulsatrix::FeedForwardModule::named_children ( )
overridevirtual

This module's direct sub-modules, each under the prefix its parameters carry in named_parameters(), in the same order (TRN-7: how ApplyLora() finds a model's Linear layers). Default: none. NamedModules() walks the whole tree.

Note
The containers that build language models and their layers report their children: SequentialModule, ResidualModule, the attention, MLP and transformer blocks, CausalLM and EncoderLM. A module that doesn't is a leaf to NamedModules().

Reimplemented from pulsatrix::Module.

◆ named_parameters()

std::vector< NamedParamRef > pulsatrix::FeedForwardModule::named_parameters ( )
overridevirtual

This module's trainable parameters, each with its hierarchical name – the one place a module declares its parameters (roadmap FND-1).

Returns
{name, {value, grad}} entries pointing directly at this module's own members, in a fixed order. Names are unique within the module tree. Default: empty (a parameterless module like ReluModule needs no override).
Note
Override this, not parameters(): saving, loading, freezing by name and optimizer parameter groups all key on these names.

Reimplemented from pulsatrix::Module.

◆ op_type()

OpType pulsatrix::FeedForwardModule::op_type ( ) const
inlineoverridevirtual

A chain of other modules.

Implements pulsatrix::Module.

◆ propagate_relevance()

Tensor pulsatrix::FeedForwardModule::propagate_relevance ( const Tensor &  relevance_out,
const LRPRuleConfig &  config 
)
overridevirtual

Computes this module's contribution to LRP relevance propagation.

Parameters
relevance_outRelevance at this module's output.
configSelects the LRP rule variant.
Returns
Relevance at this module's input.

Implements pulsatrix::Module.

◆ release_activations()

void pulsatrix::FeedForwardModule::release_activations ( )
overridevirtual

Frees the tensors forward() kept for backward() and the relevance passes; containers release their children too. Inference that only needs the output calls it after each forward, so a deep model holds one layer's activations at a time (a 1024-token ESM-2 650M pass otherwise keeps over 10 GB of them).

Note
Until the next forward(), backward() and relevance calls are invalid. The default has nothing to free.

Reimplemented from pulsatrix::Module.

◆ set_training()

void pulsatrix::FeedForwardModule::set_training ( bool  training)
overridevirtual

Sets this module's training/eval mode. Defaults to training (matches every mainstream framework's Module default).

Note
Virtual since Phase 6 Mission 6 (SequentialModule) – Mission 5 originally shipped this as plain non-virtual state ("no composite container exists yet to cascade through"); SequentialModule overrides this to cascade to every contained layer, and needs virtual dispatch to do so correctly even when accessed through a Module* base pointer, not just its own concrete type.
BatchNormModule keeps running statistics in training mode and normalizes with them in eval mode; DropoutModule is the identity in eval mode. Call set_training(false) before explaining a model that has either, so a sample's explanation doesn't depend on the rest of its batch.

Reimplemented from pulsatrix::Module.


The documentation for this class was generated from the following file: