© 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. 160280
    Prev
    Next
    INDEX
    Explanations

    end of phrase lengths

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

    length and shortening

    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
     élevé
    0.49
     élevée
    0.49
     élevées
    0.47
     높아
    0.45
     élevés
    0.45
    Ꭽ
    0.45
     cometido
    0.44
    獬
    0.44
     बीसीसीआई
    0.44
    回到
    0.42
    POSITIVE LOGITS
     Length
    0.63
     length
    0.63
     shorten
    0.61
     lengths
    0.59
     shortening
    0.59
     длина
    0.59
     shortened
    0.58
    length
    0.57
     lenght
    0.56
     short
    0.55
    Activations Density 0.076%

    No Known Activations