© 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. 10-GEMMASCOPE-RES-131K
    7. 130385
    Prev
    Next
    INDEX
    Explanations

    detract, rebuke, covariates, misplaced

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

    The neuron fires on non‐“-ing” verb stems and their inflected forms (e.g. base verbs, past participles, agent nouns).

    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
    ]
    -3.31
    ."
    -3.08
     This
    -2.73
     esteemed
    -2.61
    詳細
    -2.48
    All
    -2.47
    {
    -2.47
    re
    -2.45
    |
    -2.42
     It
    -2.41
    POSITIVE LOGITS
    
    2.61
    ⢓
    2.61
    跸
    2.59
    逦
    2.58
     zoude
    2.58
    箆
    2.52
     Preferencias
    2.44
    aciones
    2.42
     صفحۀ
    2.39
    gegangen
    2.39
    Activations Density 0.008%

    No Known Activations