© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 22458
    Prev
    Next
    INDEX
    Explanations

    Microsoft 365 user and site administration

    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/layer_40_width_262k_l0_medium
    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
     átomos
    0.43
    뮐
    0.43
    เคล
    0.42
    𒅅
    0.41
    계를
    0.41
     الاق
    0.40
     आइसलैंड
    0.40
    砒
    0.40
     ActiveX
    0.40
     রচনা
    0.39
    POSITIVE LOGITS
     conditional
    0.47
     Conditional
    0.47
    conditional
    0.46
     unconditionally
    0.44
     site
    0.41
    Bay
    0.41
    Conditional
    0.41
     chuck
    0.41
    RequestBuilder
    0.40
    Chu
    0.40
    Activations Density 0.006%

    No Known Activations