© 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. 9229
    Prev
    Next
    INDEX
    Explanations

    second, people, broken, extra

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

    The neuron fires on standout tokens such as proper names, titles, years, model numbers, and other atypical or rare “named‐entity” tokens.

    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
     hivyo
    0.84
    𝗄
    0.81
    ﻟ
    0.80
    गोरि
    0.77
    わない
    0.76
     desvi
    0.75
    ﾐ
    0.75
    徙
    0.75
    및
    0.75
     चाहिँ
    0.74
    POSITIVE LOGITS
    nél
    0.80
    dashed
    0.71
    t
    0.69
     Giá
    0.67
    TE
    0.66
     It
    0.66
    dagen
    0.66
    রে
    0.64
    college
    0.64
    wissenschaft
    0.63
    Activations Density 0.000%

    No Known Activations