© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20799
    Prev
    Next
    INDEX
    Explanations

    lava, guava, agony, lye

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

    The neuron detects any occurrence of the substring “lava” within a token, regardless of context.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    踽
    -2.75
     godk
    -2.75
    
    -2.72
     dieß
    -2.63
     メニュー
    -2.63
     dezelve
    -2.61
    銫
    -2.53
    哢
    -2.50
    􀂃
    -2.50
     بنات
    -2.48
    POSITIVE LOGITS
    扪
    2.53
     
    2.50
     interplay
    2.39
     Когда
    2.30
     Чтобы
    2.27
     "
    2.25
     influx
    2.20
     empathetic
    2.20
     meine
    2.16
    楽しみに
    2.16
    Activations Density 0.002%

    No Known Activations