© 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. 41-GEMMASCOPE-2-RES-262K
    7. 2773
    Prev
    Next
    INDEX
    Explanations

    stated rationale

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post_all/layer_41_width_262k_l0_big
    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
     correctamente
    0.56
     правильно
    0.50
     correttamente
    0.50
     correctement
    0.49
     correctly
    0.46
     properly
    0.46
     अच्छे
    0.46
    َرْ
    0.46
    effects
    0.45
     فراهم
    0.45
    POSITIVE LOGITS
     priorities
    0.65
     approach
    0.61
     emph
    0.59
     подход
    0.58
     wording
    0.57
     Priorities
    0.57
     format
    0.55
     perceptions
    0.53
    の使い方
    0.52
     how
    0.52
    Activations Density 0.004%

    No Known Activations