© 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. 65447
    Prev
    Next
    INDEX
    Explanations

    guarantee and returns

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

    The neuron detects promotional guarantee and refund/returns language (e.g., “money back guarantee,” “no questions asked,” “free shipping,” “3-day returns,” 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
     their
    -1.55
     only
    -1.43
     those
    -1.34
     more
    -1.29
     from
    -1.25
     they
    -1.24
     it
    -1.24
     his
    -1.19
     first
    -1.16
     especially
    -1.15
    POSITIVE LOGITS
     kredi
    1.08
    jón
    1.05
     SIMILAR
    1.05
    сить
    1.05
    —¿
    1.04
    рюкзак
    1.04
     AIAA
    1.02
     ------
    1.02
     MINI
    1.02
    called
    1.01
    Activations Density 0.001%

    No Known Activations