© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8992
    Prev
    Next
    INDEX
    Explanations

    .pdf URLs and filenames

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

    This neuron detects mentions of downloading or accessing documents (e.g. “Download as PDF/File,” “read online,” “free download,” file‐format names, presentation slides).

    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_31_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
    ās
    0.71
    abilă
    0.68
    ROff
    0.63
    Qs
    0.63
    َد
    0.62
    ǘ
    0.62
    āb
    0.61
    ǧ
    0.60
    ău
    0.60
    risto
    0.59
    POSITIVE LOGITS
     dyspe
    0.59
     VIT
    0.59
     приведен
    0.56
     reporte
    0.56
     bifur
    0.56
     related
    0.55
     rheumatism
    0.55
     synopsis
    0.55
     Appraisal
    0.55
     Polytechn
    0.54
    Activations Density 0.017%

    No Known Activations