| # Runnable method tutorials |
|
|
| These scripts contain the same complete workflows shown in the |
| [method guides](../../docs/methods/README.md): load a model, prepare raw inputs, |
| configure a probe, compute results, inspect tensors and draw built-in plots. |
| They do not load archived probe measurements. |
|
|
| From the repository root: |
|
|
| ```bash |
| python -m pip install -e ".[transformers,visualization]" |
| python examples/tutorials/lens_logit.py |
| ``` |
|
|
| | Family | Scripts | |
| | --- | --- | |
| | Lens | [Logit](lens_logit.py), [Tuned](lens_tuned.py), [Attention Lens](lens_attention.py), [Jacobian](lens_jacobian.py), [EmbedLens](lens_embed.py), [Patchscope](lens_patchscope.py) | |
| | Attention | [Profile](attention_profile.py), [Rollout](attention_rollout.py), [Relevance](attention_relevance.py), [Grad-CAM / ATTATTR / TAM / Beyond Intuition](attention_image_methods.py), [DTD/LRP](attention_chefer.py), [Head logit contributions](attention_head_logits.py), [Reweighting](attention_reweight.py), [Temperature](attention_temperature.py) | |
| | Causal | [Activation / attribution patching](causal_activation_patching.py), [Path patching](causal_path_patching.py), [EAP / EAP-IG](causal_eap_ig.py), [Attention knockout](causal_attention_knockout.py), [Steering / VSV](causal_steering_vsv.py) | |
|
|
| `VLM_PROBING_DEVICE` selects a device and `VLM_PROBING_OUTPUT` selects an output |
| directory. `VLM_PROBING_MODEL` optionally supplies a compatible local checkpoint; |
| otherwise each script uses its stated public model. Keep overrides within the |
| script's supported model family. Image examples expose their input URL/path and |
| prompt directly; some also accept `VLM_PROBING_IMAGE`. |
|
|
| GPT-2 is the text default, SmolVLM-500M the VLM default, and Google ViT-B/16 the |
| vision-classifier default. First use downloads weights. Dense Attention Lens |
| fitting can use several gigabytes of memory and checkpoints even for GPT-2; |
| exact Jacobians and integrated attribution require repeated backwards/forwards. |
| Read the method's cost and calibration notes before starting those scripts. |
|
|
| For a guided sequence and inline outputs, use the |
| [live notebooks](../../demos/README.md). Raw sample provenance and image rights |
| are documented there. Short training splits demonstrate the protocol; use |
| independent representative data before interpreting a fitted decoder. |
|
|
