© 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. 38942
    Prev
    Next
    INDEX
    Explanations

    formulae for solving variables

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

    The neuron strongly activates on numeric tokens or number‐like strings (integers, decimals, hex literals, negative signs, 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_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
     empez
    -1.15
    }(),
    -1.06
    wants
    -1.05
    THING
    -1.02
     descub
    -1.01
     encamin
    -1.01
     évo
    -1.01
    ihnachts
    -1.00
     comenz
    -1.00
     این
    -0.99
    POSITIVE LOGITS
     in
    1.12
     with
    1.09
     åren
    1.02
    達到
    0.97
     out
    0.94
     pudesse
    0.93
    的一個
    0.93
     slår
    0.92
     شهرهای
    0.91
     surprisingly
    0.90
    Activations Density 0.045%

    No Known Activations