© 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. 72033
    Prev
    Next
    INDEX
    Explanations

    Norm McDonald, Billygan, Not applicable

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

    The neuron fires specifically on the two‐letter token “cm.”

    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
     even
    -1.26
     faldas
    -1.16
     both
    -1.13
     which
    -1.12
     albeit
    -1.11
     entirely
    -1.09
     that
    -1.03
    ftances
    -1.02
     not
    -0.99
    troppo
    -0.99
    POSITIVE LOGITS
     băr
    1.10
    因为
    1.00
    เพราะ
    0.98
     constater
    0.93
    その後
    0.92
     beinhaltet
    0.92
    ieszka
    0.91
    第一个
    0.91
     カップ
    0.91
    ји
    0.90
    Activations Density 0.000%

    No Known Activations