© 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. 93195
    Prev
    Next
    INDEX
    Explanations

    object property names

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

    The neuron activates on numeric literal tokens (digits and numbers) in code or text.

    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
     and
    -0.93
     delicioso
    -0.91
    ricle
    -0.87
     bert
    -0.86
     robusto
    -0.85
    akade
    -0.84
     bast
    -0.83
    ayangkan
    -0.83
    面に
    -0.82
    "_
    -0.81
    POSITIVE LOGITS
     seçenek
    0.99
    hringer
    0.91
     know
    0.91
     oreilles
    0.89
     выбирать
    0.88
    ⽂
    0.88
    Ange
    0.86
    ступили
    0.86
    轰隆隆
    0.85
    ATELL
    0.85
    Activations Density 0.001%

    No Known Activations