© 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. 39-GEMMASCOPE-2-TRANSCODER-262K
    7. 1255
    Prev
    Next
    INDEX
    Explanations

    differences between assumptions and premises

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_39_width_262k_l0_small_affine
    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
    ۓ
    0.41
     questioning
    0.39
     vedit
    0.39
     Wharton
    0.38
     afge
    0.38
     científ
    0.38
     scient
    0.37
    𒈬
    0.37
     คูณ
    0.37
     соци
    0.36
    POSITIVE LOGITS
    infinite
    0.42
     tam
    0.41
    shallow
    0.41
     shallow
    0.39
    瀧
    0.38
    ட்ரோ
    0.38
     exposed
    0.37
    dihydroxy
    0.37
    ObjectID
    0.37
     bonding
    0.36
    Activations Density 0.001%

    No Known Activations