© 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-TRANSCODER-262K
    7. 13229
    Prev
    Next
    INDEX
    Explanations

    recoup costs, recover losses

    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_40_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
     Cis
    0.42
    挑
    0.42
     cis
    0.37
    ቐ
    0.37
    |^{
    0.36
     Pin
    0.36
     Marcel
    0.36
     Dale
    0.36
    _{+
    0.35
    DrawerToggle
    0.35
    POSITIVE LOGITS
     recoup
    1.45
     recuper
    1.37
     recuperar
    1.33
     recovered
    1.32
     recover
    1.29
     Recover
    1.28
     recovering
    1.27
     recou
    1.27
     recovers
    1.23
    Recover
    1.20
    Activations Density 0.021%

    No Known Activations