© 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. 22-GEMMASCOPE-RES-131K
    7. 71248
    Prev
    Next
    INDEX
    Explanations

    asking why

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

    The neuron is broadly looking for content‐bearing non–function words (i.e. salient nouns, verbs, adjectives, etc.) across lines of lyrics or poetry.

    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
     kommentar
    -0.87
     Nieren
    -0.85
     bicara
    -0.84
     sooooo
    -0.84
     —
    -0.83
     policiers
    -0.83
     kişilik
    -0.82
     kontakta
    -0.81
    wifery
    -0.81
    —¡
    -0.80
    POSITIVE LOGITS
    горе
    1.03
    生长
    0.88
     alimentaria
    0.84
    autoIncrement
    0.83
    dito
    0.82
    ueba
    0.79
     SFR
    0.79
    十分钟
    0.79
     giữ
    0.79
     seiz
    0.79
    Activations Density 0.026%

    No Known Activations