© 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. 71740
    Prev
    Next
    INDEX
    Explanations

    conjunctions and punctuation followed by various words

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

    Riemann Hypothesis, construction, file permissions

    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
     شاعر
    0.50
    .
    0.47
     سیاست
    0.46
    尽可能
    0.46
     możliwe
    0.44
    യായിരുന്നു
    0.43
     Programme
    0.42
     géographique
    0.42
     Wastewater
    0.42
    θεί
    0.42
    POSITIVE LOGITS
     tierras
    0.54
    INED
    0.53
    ecos
    0.48
     appare
    0.48
    UIActions
    0.48
    ק
    0.47
     moons
    0.47
     aparel
    0.46
     dices
    0.46
     socios
    0.46
    Activations Density 0.009%

    No Known Activations