© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 93435
    Prev
    Next
    INDEX
    Explanations

    verb verb-ending

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

    The neuron activates strongly on Japanese verb stems (the core kanji or root forms of verbs).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     with
    -1.06
     mariana
    -1.02
    Technologie
    -0.99
    لهم
    -0.97
     MISCELLANEOUS
    -0.95
     napoli
    -0.95
     Configura
    -0.94
     ecclesias
    -0.93
     marta
    -0.93
     lägen
    -0.91
    POSITIVE LOGITS
    れる
    1.04
    尨
    0.98
    する
    0.94
    aned
    0.94
    или
    0.93
    olia
    0.92
    ранее
    0.91
     três
    0.90
    ;++
    0.89
     quiero
    0.89
    Activations Density 0.002%

    No Known Activations