© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65320
    Prev
    Next
    INDEX
    Explanations

    list items and markers

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

    The neuron activates strongly on numeric tokens and list‐style markers (e.g. digits, numbered points, bullets), i.e. it detects numbers and enumerations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     участка
    -1.29
    }_{
    -1.24
     effettu
    -1.22
    _
    -1.19
     laikā
    -1.16
    TabView
    -1.15
     similar
    -1.15
    &__
    -1.13
    iacute
    -1.13
    section
    -1.11
    POSITIVE LOGITS
     favorita
    1.45
     uſ
    1.36
    ֠
    1.33
    Spesifikasi
    1.31
     newOwner
    1.28
    各种
    1.27
    Uwaga
    1.27
     cristianismo
    1.24
     grasas
    1.21
     insiste
    1.19
    Activations Density 0.033%

    No Known Activations