© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71969
    Prev
    Next
    INDEX
    Explanations

    Colorado

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

    The neuron activates on geographic location names, especially references to “Colorado” (and related place names like Denver or counties).

    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
     Çünkü
    -1.52
     статті
    -1.51
    
    -1.46
    場に
    -1.43
     Nonetheless
    -1.42
     automne
    -1.39
     causado
    -1.38
     dovuto
    -1.38
     bedeut
    -1.38
     tournoi
    -1.37
    POSITIVE LOGITS
     Colorado
    2.11
    ↵
    1.86
     was
    1.69
    ون
    1.67
     colorado
    1.66
    1.66
    Colorado
    1.60
     soal
    1.54
     {
    1.53
    Up
    1.52
    Activations Density 0.026%

    No Known Activations