INDEX

Explanations

identifying information or entities

np_acts-logits-general · gemini-2.5-flash-lite

This neuron detects occurrences of “identifiable” (especially in the context of personal data or identifiers, e.g. “personally identifiable information,” “individually identifiable,” or “online identifier”).

oai_token-act-pair · o4-mini Triggered by @jyhe0408

the word "identifiable" in privacy policy and legal contexts, particularly when preceded by "personally" or similar qualifiers.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

language referring to personal identification and unique identifiers, especially mentions of personally identifiable information in privacy or policy contexts.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ía

0.89

ний

0.85

ism

0.80

ORIAL

0.79

어

0.79

up

0.78

ualitas

0.78

евич

0.77

कर

0.75

いた

0.75

POSITIVE LOGITS

ifiable

0.67

 compi

0.64

ifiant

0.62

 danger

0.61

 nghiêm

0.61

 harm

0.61

 unambiguous

0.59

cation

0.58

мино

0.58

 sque

0.57

Activations Density 0.393%

identifying information or entities

This neuron detects occurrences of “identifiable” (especially in the context of personal data or identifiers, e.g. “personally identifiable information,” “individually identifiable,” or “online identifier”).

the word "identifiable" in privacy policy and legal contexts, particularly when preceded by "personally" or similar qualifiers.

language referring to personal identification and unique identifiers, especially mentions of personally identifiable information in privacy or policy contexts.

No Comments

No Known Activations