© 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. 34-GEMMASCOPE-RES-131K
    4. 84088
    Prev
    Next
    INDEX
    Explanations

    second nature

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

    This neuron fires on tokens that look like code/configuration directives or header‐style labels (e.g. “tslint:”, “Authorization:”, section or rule names).

    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
    retries
    -0.68
     Salazar
    -0.67
    Ад
    -0.67
     rejecting
    -0.66
    <0x9C>
    -0.66
    ီ
    -0.63
     explosión
    -0.63
    Claims
    -0.61
    摂
    -0.61
     bodied
    -0.60
    POSITIVE LOGITS
    感的
    0.73
    matian
    0.70
    postValue
    0.68
    deira
    0.68
    ясь
    0.65
     Philips
    0.65
     Ministries
    0.64
     bacterial
    0.64
     کول
    0.64
    アンス
    0.63
    Activations Density 0.072%

    No Known Activations