© 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. 31-GEMMASCOPE-2-RES-262K
    7. 24416
    Prev
    Next
    INDEX
    Explanations

    placeholders for codes and numbers

    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_262k_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
    ac
    0.58
     olacaktır
    0.53
     yardımcı
    0.52
    案内
    0.49
     undertook
    0.48
    𝚖
    0.48
     bahsede
    0.47
    𝚣
    0.47
    '],
    0.47
     avantages
    0.47
    POSITIVE LOGITS
    ש
    0.55
    ة
    0.51
    োপ
    0.46
    ोर
    0.44
     الس
    0.44
     الض
    0.44
     More
    0.44
    <0x0D>
    0.43
     Make
    0.43
     النهائي
    0.42
    Activations Density 0.006%

    No Known Activations