© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-4B-IT
    5. 5-GEMMASCOPE-2-TRANSCODER-262K
    7. 3438
    Prev
    Next
    INDEX
    Explanations

    joint probability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/transcoder_all/layer_5_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
    alyse
    1.77
     zatim
    1.71
     brilliant
    1.64
     exclamation
    1.61
     cheering
    1.59
     blanch
    1.58
     begun
    1.58
    ����
    1.58
    ures
    1.57
     dealings
    1.56
    POSITIVE LOGITS
    Swatch
    1.63
    থায়
    1.52
    irre
    1.52
    reshape
    1.51
    εις
    1.51
    िंक
    1.50
    <unused413>
    1.49
    dP
    1.48
    Exponential
    1.46
    Mét
    1.41
    Activations Density 0.000%

    No Known Activations