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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2456
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently lights up on substantive, domain-specific content words (not common function words), such as proper nouns and technical/topic terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_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
     rebuke
    0.86
     enkelt
    0.82
    TikTok
    0.80
     heartwarming
    0.80
     kiddos
    0.80
    🥹
    0.80
    ijnlijk
    0.78
    🥺
    0.77
     heartbreak
    0.77
     bustling
    0.76
    POSITIVE LOGITS
     또는
    0.93
    หรือ
    0.92
     ή
    0.85
     veya
    0.80
    または
    0.79
     or
    0.79
     или
    0.79
     හෝ
    0.77
     ಅಥವಾ
    0.76
     vagy
    0.74
    Activations Density 2.465%

    No Known Activations