© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35113
    Prev
    Next
    INDEX
    Explanations

    code identifiers and paths

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

    The neuron fires on tokens containing numbers—e.g. version numbers, numeric constants, or identifiers with digits.

    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
     libere
    -1.04
     liberi
    -1.02
     Danh
    -0.91
    Hinter
    -0.89
     konkre
    -0.88
    鲢
    -0.88
    ➹
    -0.88
     integrable
    -0.86
    を追加
    -0.86
     حتما
    -0.85
    POSITIVE LOGITS
     you
    0.84
     $\
    0.83
    ようと
    0.79
     all
    0.78
     its
    0.78
     също
    0.77
    אם
    0.77
    urtz
    0.77
    يدي
    0.76
     With
    0.75
    Activations Density 0.045%

    No Known Activations