© 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. 14673
    Prev
    Next
    INDEX
    Explanations

    going over

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

    The neuron flags very high-frequency (common) English tokens—i.e. the most frequent function words and other common words.

    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
     également
    -1.14
    にほん
    -1.06
     nuevamente
    -1.02
    osten
    -1.02
    Additionally
    -0.99
    ném
    -0.96
     indien
    -0.94
     حتی
    -0.94
    樨
    -0.93
    estern
    -0.92
    POSITIVE LOGITS
     hlavně
    1.27
     usadas
    1.22
     разных
    1.21
    けど
    1.13
     people
    1.09
     wrongly
    1.05
    olyte
    1.05
     gave
    1.01
     lately
    0.99
     usado
    0.98
    Activations Density 0.021%

    No Known Activations