© 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. 6075
    Prev
    Next
    INDEX
    Explanations

    autocomplete, picking, sexual keywords

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

    The neuron spikes on unusually rare or infrequent subword tokens (i.e. it detects tokens that occur very seldom in the text rather than common 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
    larıyla
    0.92
    Ş
    0.85
    뇰
    0.77
    IFEST
    0.76
    naden
    0.74
    IST
    0.73
    lık
    0.73
    몰
    0.71
    raft
    0.70
    rou
    0.69
    POSITIVE LOGITS
     incertid
    0.86
    避け
    0.82
     adicionar
    0.80
     sólidos
    0.80
     zahlen
    0.79
     schlecht
    0.79
     eingesetzt
    0.78
     материал
    0.77
     overgrown
    0.76
     overseen
    0.76
    Activations Density 0.001%

    No Known Activations