© 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. 34-GEMMASCOPE-2-TRANSCODER-262K
    7. 18573
    Prev
    Next
    INDEX
    Explanations

    trying to

    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_34_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
    Excellent
    0.39
     Excellent
    0.38
    鵜
    0.38
    縊
    0.36
    roid
    0.35
    ल्या
    0.34
    inned
    0.34
     übernimmt
    0.34
    ɳ
    0.34
    dess
    0.34
    POSITIVE LOGITS
     trying
    1.41
    trying
    1.27
     Trying
    1.22
     пыта
    1.21
    Trying
    1.16
     versuchen
    1.14
    试图
    1.14
     سعی
    1.14
     try
    1.13
     försö
    1.13
    Activations Density 0.151%

    No Known Activations