© 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. 193550
    Prev
    Next
    INDEX
    Explanations

    reshape, adapt, interception

    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
    ynthesis
    0.45
     severe
    0.44
    thys
    0.42
     alum
    0.41
     aerospace
    0.40
    扼
    0.40
    severe
    0.40
     behavioural
    0.39
    ถ่าย
    0.38
     localised
    0.38
    POSITIVE LOGITS
     Creatures
    0.46
    驟
    0.40
    糰
    0.39
     creature
    0.39
    移動
    0.38
     alojamiento
    0.38
     criaturas
    0.38
    Woods
    0.37
     maneuver
    0.37
     Adventures
    0.37
    Activations Density 0.002%

    No Known Activations