© 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. 31-GEMMASCOPE-2-RES-65K
    7. 30642
    Prev
    Next
    INDEX
    Explanations

    Focus on consequences and outcomes

    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/layer_31_width_65k_l0_medium
    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
     hunk
    0.41
    _
    0.41
     है
    0.40
    见
    0.39
    કની
    0.38
     Unsupported
    0.38
    稻
    0.38
    回到
    0.38
     लौ
    0.38
    不仅
    0.38
    POSITIVE LOGITS
     ovales
    0.55
    oworld
    0.45
     interessant
    0.43
     trataro
    0.43
     salari
    0.42
     logros
    0.42
     VO
    0.42
    ボリー
    0.42
     freund
    0.41
     WOW
    0.41
    Activations Density 0.020%

    No Known Activations