© 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. 13204
    Prev
    Next
    INDEX
    Explanations

    usually contrasted

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

    The neuron spikes on very common, high‐frequency words (e.g. “it,” “’s,” “a,” “the,” etc.).

    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
    etta
    -0.90
     вместо
    -0.89
    <eos>
    -0.85
    MeasureSpec
    -0.83
     religi
    -0.81
    Bukan
    -0.81
    Doch
    -0.80
    viamente
    -0.80
    ismus
    -0.79
    蛧
    -0.79
    POSITIVE LOGITS
     relativamente
    1.10
    porre
    1.04
    uaa
    1.03
    tares
    1.02
    olesale
    0.95
    vimos
    0.94
    словно
    0.94
     enemmän
    0.93
    cknowled
    0.91
     klikken
    0.91
    Activations Density 0.145%

    No Known Activations