© 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. 34-GEMMASCOPE-RES-131K
    7. 79060
    Prev
    Next
    INDEX
    Explanations

    programming code and configuration

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

    The neuron fires on numeric tokens—i.e. literal numbers (constants, version numbers, digits in identifiers, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    italien
    -0.99
    Esperamos
    -0.95
     renon
    -0.91
     funerals
    -0.91
    ֍
    -0.90
     perspectiva
    -0.90
     avete
    -0.89
    Excited
    -0.89
    realpath
    -0.88
     sanctified
    -0.88
    POSITIVE LOGITS
     his
    1.07
     its
    0.99
    :][
    0.95
     vaccine
    0.93
     jego
    0.90
     exam
    0.90
     его
    0.89
    },
    0.89
     project
    0.86
     app
    0.84
    Activations Density 0.078%

    No Known Activations