Research model representations
Skill gaelic-ghost/socket/skills/research-model-representations
Design causal research into model activations, features, attention, residual streams, probes, and circuits. Use when locating behavior, testing a direction or feature, comparing layers, or reproducing interpretability research.From its SKILL.md
npx -y skills add gaelic-ghost/socket --skill research-model-representationsAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.4 KB, 397 tokens by cl100k_base, as published. Nobody here has run it
Research Model Representations
Start With A Causal Question
Name the behavior, candidate representation, intervention point, predicted behavioral change, and falsifying result. A decodable probe or correlated activation is evidence of information, not evidence that the model uses it causally.
Workflow
- Pin the checkpoint, tokenizer, template, framework, and source-code revision.
- Construct matched positive, negative, and neutral examples. Control length, topic, syntax, and token positions where practical.
- Define hook names by verified model architecture rather than assuming layer paths from a related model.
- Decide local versus remote instrumentation. Before remote execution, classify prompts and activations, confirm authorization, retention/logging behavior, network exposure, and paid-compute budget; keep proprietary or sensitive inputs local unless explicitly approved.
- Collect activations with explicit batch, dtype, device, token-selection, pooling, and normalization rules.
- Split probe training and evaluation examples by the true contamination boundary.
- Establish selectivity controls: random labels, random directions, held-out concepts, and simple surface-feature baselines.
- Test causality through ablation, patching, steering, or counterfactual replacement at held-out examples.
- Measure target behavior and unrelated capability guardrails across layers, positions, and intervention strengths.
- Report unstable seeds, negative results, multiple-comparison choices, and architecture-specific limitations.
Tool Selection
- TransformerLens provides activation caching and hook-oriented analysis for supported architectures.
- NNsight provides model instrumentation and remote-capable intervention workflows.
Verify the pinned model and operation against the selected tool. Use direct framework hooks when a small, explicit intervention is clearer than adding a large abstraction.
References
Read references/causal-representation-evidence.md before interpreting a probe or direction as a mechanism.
What ships with it: 2 files
1.1 KB alongside SKILL.md
agents/
- openai.yaml264 B