© 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. 26219
    Prev
    Next
    INDEX
    Explanations

    specifying what follows

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

    This neuron detects explicit correction or revision directives in technical documents, such as phrases like “should be removed and replaced with the following:”.

    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
    digheden
    -1.01
    матика
    -0.96
    
    -0.95
    qxd
    -0.94
     préférable
    -0.93
    gın
    -0.91
    APER
    -0.91
     héro
    -0.91
     septembrie
    -0.91
     privilégi
    -0.90
    POSITIVE LOGITS
    กรณ์
    0.94
     появится
    0.89
    <<<<<<<
    0.86
    propy
    0.84
     with
    0.84
     künftig
    0.84
     watermark
    0.84
    ,'\
    0.83
     newfound
    0.83
     nowej
    0.81
    Activations Density 0.012%

    No Known Activations