© 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. 72830
    Prev
    Next
    INDEX
    Explanations

    thank you

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

    The neuron fires on standalone integer tokens that represent comment or post score counts (e.g. “157,” “342,” “245,” etc.).

    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
     japonais
    -1.34
    vård
    -1.27
     istimewa
    -1.26
     bicara
    -1.25
     signifikan
    -1.24
     légales
    -1.24
     koneksi
    -1.24
     Karakter
    -1.23
    Dzi
    -1.22
     tém
    -1.21
    POSITIVE LOGITS
     they
    1.41
    can
    1.38
     see
    1.35
     anything
    1.25
    no
    1.22
     !
    1.20
    ↵
    1.19
    などを
    1.19
     wasn
    1.19
     always
    1.18
    Activations Density 0.005%

    No Known Activations