© 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. 81090
    Prev
    Next
    INDEX
    Explanations

    business, power, trash, fear, India, size

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

    grammar, attention, capture, virtualization

    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
    adlo
    0.49
    áll
    0.46
     አለበት
    0.46
     heartily
    0.45
     stifle
    0.44
    រួ
    0.44
    基本的に
    0.43
    苇
    0.43
    规格
    0.42
    部分的
    0.42
    POSITIVE LOGITS
     CJ
    0.48
     Carg
    0.45
     sabato
    0.44
     Predictive
    0.43
     WJ
    0.43
     Führung
    0.42
     Goll
    0.41
    CJ
    0.41
    𝗖
    0.41
     porówn
    0.41
    Activations Density 0.000%

    No Known Activations