© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13397
    Prev
    Next
    INDEX
    Explanations

    variables assigned values

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

    This neuron flags salient action or advisory keywords—verbs and nouns used in instructions, procedures, or cautionary/advice contexts.

    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
    ਿਸ
    0.95
    ନା
    0.81
    ਾਲ
    0.80
     д
    0.78
    不得不
    0.77
     Ш
    0.77
    <0x96>
    0.73
    ੂ
    0.72
     Г
    0.71
    <0x94>
    0.70
    POSITIVE LOGITS
    ateboard
    0.88
    Wat
    0.84
    GetAll
    0.79
    たちは
    0.79
     piatti
    0.77
     espectáculo
    0.76
    Chat
    0.75
    砾
    0.75
    発電
    0.74
     snowboard
    0.74
    Activations Density 0.000%

    No Known Activations