© 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. 196048
    Prev
    Next
    INDEX
    Explanations

    Non-English words such as 'hat', 'ऐसे', 'hamil', 'vai'

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

    suggestions or clarifying context

    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
    ن
    0.57
    сову
    0.55
     ዲ
    0.49
     ይታ
    0.47
    ল
    0.46
    ോ
    0.46
     giriyoruz
    0.44
    νά
    0.44
    става
    0.43
    сова
    0.43
    POSITIVE LOGITS
    0.57
     hat
    0.54
     ऐसे
    0.52
    hat
    0.50
     hamil
    0.50
     vai
    0.48
     $\
    0.48
     nhanh
    0.47
     wynik
    0.47
     that
    0.47
    Activations Density 0.000%

    No Known Activations