© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 9374
    Prev
    Next
    INDEX
    Explanations

    questions and ecotoxicology

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

    code block closing brackets or parentheses, especially at the end of JavaScript/CSS functions.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron flags numeric constants (literal numbers) in the code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     къде
    -1.20
    頹
    -1.14
    ových
    -1.10
     évé
    -1.09
    気持ちが
    -1.04
    räck
    -1.02
    bahan
    -1.02
    っちゃう
    -1.02
    忘記
    -1.01
     artikelen
    -1.00
    POSITIVE LOGITS
     a
    1.28
     estructuras
    1.19
    ];
    1.11
    mer
    1.07
     promesas
    1.05
     propuestas
    1.04
    return
    1.03
    ệu
    1.02
     cualidades
    1.00
    жнему
    1.00
    Activations Density 0.003%

    No Known Activations