© 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. 3008
    Prev
    Next
    INDEX
    Explanations

    organizations, funds, or communities

    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
     schematically
    1.85
    мам
    1.82
    tion
    1.80
    tı
    1.79
    𝐭
    1.79
    てる
    1.78
    ލ
    1.78
    Thats
    1.73
    Wouldn
    1.73
     мероприятия
    1.71
    POSITIVE LOGITS
    л
    1.65
    ع
    1.61
     odnos
    1.44
     urb
    1.44
     evil
    1.40
    ⎯
    1.38
    s
    1.37
    ਆਂ
    1.37
    QUARE
    1.37
    ே
    1.35
    Activations Density 0.000%

    No Known Activations