© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100370
    Prev
    Next
    INDEX
    Explanations

    relation between items

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

    This neuron detects prepositions (particularly “on” or “in”) when they introduce or precede a numeric quantity.

    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
     that
    -2.09
     whose
    -1.79
     out
    -1.64
     spurred
    -1.63
     also
    -1.57
     epitom
    -1.55
    𝑧
    -1.55
     this
    -1.54
    enfone
    -1.47
     keeping
    -1.45
    POSITIVE LOGITS
    3
    1.92
    镲
    1.59
    three
    1.56
     طريقة
    1.52
     declaração
    1.52
    five
    1.51
     propone
    1.50
     importância
    1.50
     ,‬
    1.50
    ){
    1.49
    Activations Density 0.136%

    No Known Activations