© 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. 34-GEMMASCOPE-RES-131K
    7. 32286
    Prev
    Next
    INDEX
    Explanations

    represented by, nil pointer

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

    This neuron activates on unusually long tokens (typically words or numbers much longer than average).

    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
    ylation
    -0.85
     entityManager
    -0.80
     window
    -0.80
    schema
    -0.80
    yards
    -0.77
    之一
    -0.77
    itals
    -0.76
     appContext
    -0.75
    upply
    -0.75
     instructions
    -0.73
    POSITIVE LOGITS
     exposé
    0.77
     minibar
    0.76
     antiseptic
    0.75
    ὅ
    0.75
    カテゴリ
    0.75
     Techn
    0.74
     xenon
    0.73
     anticor
    0.72
    literature
    0.71
    ואל
    0.71
    Activations Density 0.009%

    No Known Activations