© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 66813
    Prev
    Next
    INDEX
    Explanations

    business transactions and results

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

    disThis neuron activates for various types of text, including actions like 'remove', 'dropping', 'defining', 'prioritizing', 'surprise' followed by specific subsequent words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Char
    0.46
     często
    0.41
     Bj
    0.40
     často
    0.39
     older
    0.38
     Altos
    0.38
     varchar
    0.37
     Β
    0.37
     많은
    0.37
     Gardner
    0.37
    POSITIVE LOGITS
    finale
    0.56
    ؟
    0.53
    ؟!
    0.48
    nél
    0.48
    ؟؟
    0.48
    ruptcy
    0.47
     tikai
    0.47
    结局
    0.47
    seater
    0.46
    essentially
    0.46
    Activations Density 0.005%

    No Known Activations