© 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. 13776
    Prev
    Next
    INDEX
    Explanations

    optimal x, whereby

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

    The neuron strongly fires on unusually large runs of whitespace or other odd formatting tokens (e.g. many consecutive spaces).

    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
     બ્
    0.85
     contribu
    0.84
     スペ
    0.84
     テーブル
    0.78
    ́p
    0.76
    ভূম
    0.75
    殡
    0.75
    ூன்
    0.73
     キャップ
    0.73
     メガネ
    0.73
    POSITIVE LOGITS
    seits
    0.84
    theless
    0.81
    േരി
    0.79
     zich
    0.77
    ле
    0.76
    essentially
    0.75
    能否
    0.73
    是否
    0.73
    謂
    0.73
    телей
    0.72
    Activations Density 0.000%

    No Known Activations