© 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. 4633
    Prev
    Next
    INDEX
    Explanations

    `March` `tips`

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

    The neuron strongly activates on rare, content‐heavy words (e.g. proper names and technical or domain‐specific terms) rather than common function words.

    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
    Activation
    0.73
    種
    0.70
     eben
    0.70
    潴
    0.70
     этим
    0.68
    仅仅
    0.68
    Revision
    0.68
    clature
    0.67
     nourished
    0.66
    Rotation
    0.66
    POSITIVE LOGITS
    ratto
    0.82
     eventuali
    0.79
     돼
    0.76
     strictest
    0.75
     vagin
    0.75
     Ofer
    0.75
     опла
    0.74
     educativa
    0.73
     собой
    0.73
     referendum
    0.72
    Activations Density 0.001%

    No Known Activations