© 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. 128106
    Prev
    Next
    INDEX
    Explanations

    Explanation of neuron 4 behavior: the main thing this neuron does is find strings of digits (numerical tokens).

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

    @rem

    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
     banyo
    -1.20
     arka
    -1.17
     senin
    -1.16
    moiselle
    -1.15
     arkas
    -1.13
    থ্য
    -1.13
     boxe
    -1.10
     italiano
    -1.10
     ché
    -1.10
     golos
    -1.09
    POSITIVE LOGITS
    J
    1.20
    ik
    1.20
    C
    1.18
    im
    1.16
    il
    1.14
    F
    1.09
    E
    1.09
    B
    1.09
    炜
    1.07
    M
    1.05
    Activations Density 0.000%

    No Known Activations