© 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. 5176
    Prev
    Next
    INDEX
    Explanations

    quote followed by 's'

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

    The neuron fires on technical specification tokens—especially alphanumeric codes, version numbers, and file‐format or model identifiers (e.g. “mp3,” “wav,” “WMA,” “Act 2,” “3.8.2,” sensor specs, etc.).

    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
     وە
    0.55
     सौंप
    0.54
    <unused13>
    0.52
     кеңсеси
    0.52
    <unused75>
    0.52
     اہلِ
    0.51
    類の
    0.51
    កាន់
    0.51
    菓
    0.50
    გუ
    0.49
    POSITIVE LOGITS
     quando
    0.56
     paragraphe
    0.53
    тов
    0.51
     cuando
    0.50
     patip
    0.50
     because
    0.50
    icycles
    0.49
     porque
    0.49
    0.49
    ,@
    0.47
    Activations Density 0.962%

    No Known Activations