© 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-IT
    5. 41-GEMMASCOPE-2-RES-262K
    7. 31820
    Prev
    Next
    INDEX
    Explanations

    let's be honest

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post_all/layer_41_width_262k_l0_big
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ایکسپریس
    0.44
     によ
    0.38
    ijl
    0.35
    եք
    0.34
    #$%
    0.34
    韬
    0.34
     চূ
    0.34
    rceil
    0.34
    冝
    0.33
    zenesulf
    0.33
    POSITIVE LOGITS
     honestly
    2.30
     Honestly
    1.96
     frankly
    1.93
    Honestly
    1.83
     tbh
    1.70
     truthfully
    1.63
     Frankly
    1.59
     honest
    1.56
    正直
    1.54
     ehrlich
    1.45
    Activations Density 0.001%

    No Known Activations