© 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. 34-GEMMASCOPE-RES-131K
    7. 89766
    Prev
    Next
    INDEX
    Explanations

    came in

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

    The neuron is essentially triggered by the word “cities,” i.e. it detects mentions of cities.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    spath
    -0.79
    vrons
    -0.79
     brano
    -0.76
    НК
    -0.75
    ErrUnexpected
    -0.75
    ☦
    -0.75
    ymi
    -0.74
    )))
    -0.73
     なり
    -0.73
     repente
    -0.71
    POSITIVE LOGITS
     clock
    3.98
    clock
    3.47
     clocks
    3.45
    Clock
    3.11
     clocked
    3.08
     Clock
    2.94
     Clocks
    2.53
     CLOCK
    2.50
    CLOCK
    2.28
     coming
    1.84
    Activations Density 0.053%

    No Known Activations