© 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. 56-GEMMASCOPE-2-TRANSCODER-262K
    7. 20262
    Prev
    Next
    INDEX
    Explanations

    being or spirits

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_56_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
     ch
    3.02
     Ch
    1.86
     च
    1.85
     چ
    1.80
    ch
    1.65
    Ch
    1.63
     চ
    1.59
     ч
    1.46
    ช
    1.41
     CH
    1.11
    POSITIVE LOGITS
     spirits
    1.37
     Spirits
    1.20
    spir
    0.75
     espí
    0.62
     Spir
    0.46
     Espí
    0.45
     beings
    0.40
     spir
    0.38
    QUIN
    0.38
    Characters
    0.38
    Activations Density 0.049%

    No Known Activations