© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11449
    Prev
    Next
    INDEX
    Explanations

    readings, forecasts, technical terms

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

    The neuron chiefly fires on words that name or signal time‐related events or states (e.g. future, dates, release, record, situation).

    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
     attacked
    0.75
     lips
    0.75
     by
    0.73
    =
    0.72
     asked
    0.70
     indicates
    0.70
     et
    0.69
     called
    0.69
     byen
    0.69
     transcripts
    0.69
    POSITIVE LOGITS
     தற்போது
    0.82
    暦
    0.74
    Thankfully
    0.73
    Edad
    0.72
    はもちろん
    0.71
    ♕
    0.71
    福祉
    0.70
    に関しては
    0.70
     спустя
    0.70
    рафон
    0.69
    Activations Density 0.042%

    No Known Activations