© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 154164
    Prev
    Next
    INDEX
    Explanations

    conditional sentences

    np_acts-logits-general · gemini-2.5-flash-lite

    emphasis and negation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     الاي
    0.47
     پایه
    0.39
    ায়ন
    0.37
     QM
    0.37
    下来的
    0.36
     AY
    0.36
    門
    0.35
     lh
    0.34
     hors
    0.34
     ત્રણ
    0.34
    POSITIVE LOGITS
     inferiority
    0.46
     impedir
    0.46
     inim
    0.44
     paradoxical
    0.43
     imposed
    0.42
    entionally
    0.42
     begr
    0.41
    Conceptual
    0.41
    Suk
    0.41
     নিষেধ
    0.40
    Activations Density 0.007%

    No Known Activations