© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6330
    Prev
    Next
    INDEX
    Explanations

    `[:]` or `(tokens)`

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

    This neuron detects long numeric sequences (multi-digit numbers).

    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
     processions
    0.48
     fragmentation
    0.43
     progressivement
    0.43
    ேயே
    0.42
    $)$.
    0.42
     orchestration
    0.42
     kanggo
    0.41
     отдельные
    0.40
    ŗ
    0.39
    퐫
    0.39
    POSITIVE LOGITS
    ，
    0.47
     Meme
    0.46
    ത്തിന്
    0.46
    ahue
    0.46
     Florida
    0.44
     Incredible
    0.44
     encuentra
    0.44
    ぐらい
    0.44
     भयानक
    0.43
     Southwest
    0.43
    Activations Density 1.645%

    No Known Activations