© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8026
    Prev
    Next
    INDEX
    Explanations

    input processing

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

    The neuron primarily detects mentions of “input signals,” i.e. occurrences of the phrase input signals.

    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_31_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
    何况
    0.68
    Ral
    0.66
    Think
    0.62
    istit
    0.61
    jq
    0.61
    Aer
    0.61
    别
    0.61
     Вікіпедії
    0.60
    程度上
    0.59
    К
    0.59
    POSITIVE LOGITS
     getInput
    0.96
    ting
    0.80
    getInput
    0.78
     Input
    0.77
    setInput
    0.76
     newInput
    0.75
     INPUT
    0.73
     Inputs
    0.70
    पुट
    0.66
     输入
    0.65
    Activations Density 0.182%

    No Known Activations