Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    were

    np_max-act · gemini-2.0-flash

    the neuron detects tokens in subjunctive conditional clauses (e.g. the “if … were to be” construction).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    evidence of concerns regarding democratic processes and governance.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     bad
    -0.08
     japan
    -0.07
    Self
    -0.07
    AILS
    -0.07
     Laser
    -0.07
     PM
    -0.07
    ิส
    -0.06
     breaks
    -0.06
     defence
    -0.06
     tjejer
    -0.06
    POSITIVE LOGITS
     Were
    0.07
    อกจาก
    0.07
    ure
    0.07
    istrate
    0.07
    going
    0.07
     invitations
    0.07
    ute
    0.07
     were
    0.06
    edn
    0.06
     vere
    0.06
    Act Density 0.010%

    No Known Activations