© 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. 24-GEMMASCOPE-2-RES-16K
    7. 14580
    Prev
    Next
    INDEX
    Explanations

    resolves to, estimate the

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

    This neuron responds to uncommon, domain-specific content words (e.g. technical or specialized terms).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the modal verb "could" in contexts describing potential actions or outcomes.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    mentions of time and scheduling—dates, days, and temporal markers indicating when events occur or are completed.

    oai_token-act-pair · gpt-5Triggered 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
     সম্ভবনা
    0.72
    வைகள்
    0.71
     deoarece
    0.70
    reements
    0.68
    SourceRequest
    0.68
     পুনরায়
    0.68
     தொடங்க
    0.68
    像是
    0.67
    IONES
    0.67
     cambiamenti
    0.67
    POSITIVE LOGITS
     (
    0.89
    /
    0.75
     (!)
    0.74
     (!
    0.73
     superb
    0.73
     whatever
    0.68
     notwithstanding
    0.68
     extin
    0.66
    (!)
    0.66
     unremarkable
    0.66
    Activations Density 0.049%

    No Known Activations