© 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. 11170
    Prev
    Next
    INDEX
    Explanations

    Confidence probability

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

    The neuron is keyed to uncommon, content-bearing tokens (i.e. rare or high-information 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
    ie
    1.20
    er
    1.19
    al
    1.14
    ol
    1.08
    dings
    0.96
    an
    0.95
    ic
    0.93
    u
    0.93
    iu
    0.90
     nowDate
    0.86
    POSITIVE LOGITS
    GK
    0.74
    ِ
    0.71
     Backup
    0.69
    更好地
    0.68
    ்க
    0.67
     preserv
    0.67
    ဢ
    0.66
     Outcome
    0.66
    Munic
    0.65
     pepp
    0.65
    Activations Density 0.000%

    No Known Activations