© 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. 25854
    Prev
    Next
    INDEX
    Explanations

    measurements and quantities

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

    The neuron is effectively inactive (“dead”)—it doesn’t respond to any tokens.

    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
     new
    -1.13
     this
    -0.99
     showing
    -0.98
     which
    -0.98
     at
    -0.95
     still
    -0.91
     APPELLATE
    -0.90
     there
    -0.88
    BackPressed
    -0.87
    نامج
    -0.87
    POSITIVE LOGITS
    すっ
    1.20
     indescri
    1.03
    粜
    1.00
    很想
    0.99
    ційний
    0.98
    色々と
    0.98
    マジで
    0.98
    radura
    0.97
    ふと
    0.97
     тщательно
    0.96
    Activations Density 0.036%

    No Known Activations