© 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. 64765
    Prev
    Next
    INDEX
    Explanations

    0s, ’s, ’ve

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

    The neuron fires on numeric tokens (multi‐digit numbers and numeric expressions).

    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
    depuis
    -1.49
     captiv
    -1.42
    튿
    -1.29
     paramètres
    -1.28
     essentiels
    -1.27
    力
    -1.26
    を行いました
    -1.26
    辚
    -1.21
    گاهی
    -1.20
    へ行
    -1.20
    POSITIVE LOGITS
     and
    2.14
    :
    1.44
     with
    1.35
    '
    1.34
    <eos>
    1.33
     (
    1.27
    wester
    1.26
    panes
    1.26
     or
    1.25
     for
    1.24
    Activations Density 0.024%

    No Known Activations