© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 12190
    Prev
    Next
    INDEX
    Explanations

    adding VPN or diffuse idiopathic

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

    computer code or technical documentation elements.

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

    The neuron activates on adjective tokens (words tagged with “_JJ”), i.e. it detects adjectives.

    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
    женных
    -0.89
    蒴
    -0.83
    人了
    -0.81
    𝘖
    -0.76
     investigations
    -0.74
    dependence
    -0.74
     articula
    -0.73
     coti
    -0.73
    riad
    -0.73
    level
    -0.72
    POSITIVE LOGITS
    ("}");
    0.92
    inot
    0.91
     เพ
    0.84
    sili
    0.82
    ;}
    0.82
     също
    0.81
     mieste
    0.79
     Tarjeta
    0.77
    ]^{
    0.77
     spokoj
    0.77
    Activations Density 0.002%

    No Known Activations