© 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. 22-GEMMASCOPE-2-TRANSCODER-262K
    7. 43196
    Prev
    Next
    INDEX
    Explanations

    5-4-3-2-1 grounding technique

    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_22_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
     ಹೇಗೆ
    0.46
     linéaires
    0.43
    PHPExcel
    0.42
     tradizionale
    0.42
    ചന
    0.41
     způso
    0.41
    椋
    0.40
     பல்கலை
    0.40
    Probe
    0.40
     полости
    0.39
    POSITIVE LOGITS
    }_
    0.45
     kicking
    0.45
     presum
    0.42
    ermal
    0.42
    érm
    0.42
     Herm
    0.41
    oil
    0.41
    inke
    0.39
     odmah
    0.39
    novel
    0.38
    Activations Density 0.001%

    No Known Activations