© 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. 67906
    Prev
    Next
    INDEX
    Explanations

    names following prepositions

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

    The neuron predominantly picks up on three-digit numeric tokens (e.g. seed or ranking numbers).

    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
    traditional
    -0.94
     obdob
    -0.90
    Ḍ
    -0.90
     adds
    -0.85
     subtly
    -0.84
     dropped
    -0.84
     virtually
    -0.84
    ksiä
    -0.84
     traditional
    -0.83
    راط
    -0.83
    POSITIVE LOGITS
    OPER
    0.96
     melainkan
    0.95
     néanmoins
    0.90
    PutPixel
    0.87
     marseille
    0.84
     sorento
    0.84
     poff
    0.84
     erin
    0.84
     speciale
    0.83
     capace
    0.82
    Activations Density 0.002%

    No Known Activations