© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-1B
    5. 17-GEMMASCOPE-2-RES-16K
    7. 1585
    Prev
    Next
    INDEX
    Explanations

    who, whom, whose, quien

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

    terms that denote people—especially gendered or age-specific nouns and role titles referring to a person.

    oai_token-act-pair · gpt-5Triggered by @smilgaveronika
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-1b-pt/resid_post/layer_17_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dotycz
    1.02
     dotyczą
    1.02
     включают
    1.00
    部位
    1.00
    箇所
    0.99
    ときは
    0.99
    نان
    0.99
    곳
    0.98
     strutture
    0.98
    生效
    0.97
    POSITIVE LOGITS
     who
    1.68
    who
    1.56
     whom
    1.49
    whom
    1.44
     whos
    1.35
     extraordinaire
    1.30
     whose
    1.27
     wann
    1.25
    whose
    1.24
     born
    1.20
    Activations Density 0.535%

    No Known Activations