© 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. 128752
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on tokens that indicate removing, dropping, or closing something (i.e. deletion/removal operations).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    remove or delete

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     also
    -1.02
     MISSING
    -1.00
    Rese
    -0.91
    ==>
    -0.88
     نیز
    -0.88
     POSSIBLE
    -0.84
    ריאה
    -0.84
     осо
    -0.84
     DERE
    -0.82
     亜
    -0.78
    POSITIVE LOGITS
     unnecessary
    1.14
     inneces
    1.06
     лиш
    1.02
    mbggenerated
    0.91
     оці
    0.91
    ził
    0.91
     eigentlich
    0.90
     unwanted
    0.88
     superfluous
    0.88
    ZIONE
    0.86
    Activations Density 0.012%

    No Known Activations