© 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. 26151
    Prev
    Next
    INDEX
    Explanations

    decrease, lower, smaller

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

    The neuron is triggered by tokens that denote decreasing or “less‐than” relations—e.g. words like “decreasing” or “decreased,” the minus sign, and comparison symbols such as “<.”

    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
    toUpperCase
    -0.87
     Ascension
    -0.84
     resurrection
    -0.84
    onResume
    -0.84
    ++
    -0.81
    crtc
    -0.81
     uplift
    -0.80
    Resume
    -0.80
    increased
    -0.79
     up
    -0.79
    POSITIVE LOGITS
     down
    1.48
    down
    1.40
     lowering
    1.10
     пони
    1.03
     پایین
    1.03
     lowered
    1.02
     menurun
    1.02
     penurunan
    1.01
     вниз
    1.01
    downarrow
    1.01
    Activations Density 0.137%

    No Known Activations