© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 129436
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens—especially decimal and integer stats—i.e. the numbers reporting players’ performance figures.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    average stats and records

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     until
    -1.22
     but
    -1.16
    但在
    -1.11
     Bundesrepublik
    -1.06
    市镇
    -1.06
     of
    -1.02
    })$
    -0.99
    ꒱
    -0.99
     reveló
    -0.99
    نسية
    -0.98
    POSITIVE LOGITS
     problemy
    1.36
    verages
    1.17
     better
    1.16
     nadru
    1.15
     Czę
    1.14
     Projektu
    1.14
     average
    1.13
     hoge
    1.09
    higher
    1.08
     lakier
    1.08
    Activations Density 0.005%

    No Known Activations