© 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. 8810
    Prev
    Next
    INDEX
    Explanations

    initial consonant clusters

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

    st, tr, ch, th, br prefixes

    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
    萛
    0.39
    సూరు
    0.38
     showrooms
    0.37
    ährt
    0.37
    াঙ্গীর
    0.36
    смотря
    0.35
    MyToolkit
    0.35
    Manu
    0.35
    ဳ
    0.35
    పథ
    0.34
    POSITIVE LOGITS
    t
    0.61
    c
    0.54
    ت
    0.49
    s
    0.46
    un
    0.45
    ת
    0.45
    т
    0.43
    د
    0.43
    um
    0.42
    0.40
    Activations Density 0.056%

    No Known Activations