© 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. 13019
    Prev
    Next
    INDEX
    Explanations

    increase and decrease

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

    The neuron activates primarily on numeric values and measurements (digits, figures, and units).

    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
     Applies
    -0.81
    ępnie
    -0.79
    แค่
    -0.77
    faatkan
    -0.75
    mögen
    -0.75
     even
    -0.75
     already
    -0.74
     Later
    -0.74
     タイヤ
    -0.74
     locker
    -0.73
    POSITIVE LOGITS
    Conversely
    1.53
     conversely
    1.41
     Conversely
    1.34
     opposite
    1.26
    opposite
    1.19
     противополо
    1.16
     correspondingly
    1.10
     Opposite
    1.04
    Opposite
    1.02
     inversely
    0.99
    Activations Density 0.100%

    No Known Activations