liveBook · Manning

Activation capture

As explained in Rearchitecting LLMs

Activation capture records intermediate neural signals produced during model execution.

Activation capture records intermediate neural signals during a model’s forward pass. Hooks make it possible to inspect individual transformer submodules or collect activation statistics for data-driven neuron selection. section 5.2.3 section 10.1

Choose the capture point according to the signal needed. A forward hook records a module’s output; a forward pre-hook records its input. For neuron-level MLP analysis, capture the input of down_proj: it contains the gated, expanded signal before down_proj mixes and compresses it. In Llama-3.2-1B, down_proj maps 8,192 features to 2,048, so its output no longer preserves the same per-neuron traceability. section 5.2 section 10.1

A practical capture workflow:

  • Inspect the model hierarchy and select submodules and input or output hook points. Module names differ across architectures; a hook targeting self_attn can silently miss Qwen3.5’s linear_attn. section 10.1
  • Register hooks for the selected layers. For prompt-level inspection, detach captured tensors and move them to CPU; handle tuple outputs by selecting the desired tensor. section 10.1
  • Run the model without gradients. For calibration, use evaluation mode and process batches on the model’s device; hooks can compute L2 norms at each down_proj input and accumulate them on CPU. section 5.2.3
  • Remove every hook after capture. A finally block ensures cleanup even if the forward pass fails. On a memory-constrained GPU, periodic cache clearing can reduce VRAM pressure at a speed cost. section 10.1 section 5.2.3
Listing 10.3 Hook cleanup and activation capture
def remove_hooks(handles):
    for handle in handles:
        handle.remove()
 
 
def get_prompt_activations(model, tokenizer, prompt, target_layers=None):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    handles = register_hooks(model, 
        target_layers=target_layers) #A
    Try: #1
        with torch.no_grad():
            model(**inputs)
        return {k: v.clone() for k, v 
                in model._activations.items()} #B
    finally:
        remove_hooks(handles)
        if hasattr(model, "_activations"):
            del model._activations

The prompt-capture workflow registers hooks, runs without gradients, copies the results, and guarantees cleanup. section 10.1

Calibration for width pruning

Run calibration batches from WikiText-2 or SMS Spam through the model while hooks collect per-layer down_proj input norms. Combine the dynamic activation signal with normalized weight-range scores from gate_proj, up_proj, and down_proj. Multiplying the structural and activation scores lowers the importance of a neuron with large weights but near-zero activation, making it a pruning candidate. section 5.2.3 section 5.2

Figure 5.5 Hybrid importance calculation in GLU architecture. The static component combines weight ranges from gate_proj, up_proj, and down_proj. The dynamic component captures activation norms at the input of down_proj, where gated information flows. Both components multiply to produce the final importance score.

The diagram locates capture before down_proj and shows how activation norms combine with weight-based scores. section 5.2

How Rearchitecting LLMs covers it

How deep it goes
Walks through hook registration, calibration, CPU accumulation, and cleanup, including module-name and GPU-memory limitations.
Examples
Uses Llama-3.2-1B capture points and calibration batches from WikiText-2 and SMS Spam.
What you can do after
Apply it via PyTorch hooks to capture prompt activations or collect per-layer calibration norms.
How it connects
Connects activation capture to forward hooks, transformer MLPs, L2 norms, and width pruning.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Rearchitecting LLMs. The section links point to the parts of the book it draws on.