© 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. 137332
    Prev
    Next
    INDEX
    Explanations

    Read the full story

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

    reproducibility

    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
     DURING
    0.51
     Bengaluru
    0.49
     stimuli
    0.49
    OLINA
    0.49
    কেন
    0.47
     defam
    0.46
    <0x80>
    0.46
     рэгістра
    0.46
     NRA
    0.46
     registrazione
    0.45
    POSITIVE LOGITS
    OwnProperty
    0.51
     veces
    0.49
    MSBuild
    0.47
    StackTrace
    0.44
     houden
    0.43
    Edge
    0.42
    _+
    0.42
    Continental
    0.41
    Raised
    0.41
     rarement
    0.41
    Activations Density 0.000%

    No Known Activations