© 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. 40-GEMMASCOPE-2-RES-262K
    7. 221080
    Prev
    Next
    INDEX
    Explanations

    not mandatory

    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
    没有任何
    0.45
    Desired
    0.43
     undesirable
    0.42
     ønsker
    0.41
     desirable
    0.41
    峈
    0.40
    毫无
    0.40
    欲
    0.40
     идеа
    0.39
    कोई
    0.39
    POSITIVE LOGITS
     mandatory
    0.94
     Mandatory
    0.85
     compulsory
    0.81
     obligatoire
    0.77
     obligatory
    0.75
     necessity
    0.75
     obligatorio
    0.74
     essential
    0.73
     обяза
    0.73
     अनिवार
    0.69
    Activations Density 0.029%

    No Known Activations