© 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. 71926
    Prev
    Next
    INDEX
    Explanations

    adopted criteria, evaluation, or conditions

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

    The neuron fires on ordinary running‐text words in natural‐language prose (and stays off in code, formulas, or structured metadata).

    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
     énorm
    -1.03
    Interess
    -0.99
    Или
    -0.99
    以上です
    -0.96
    ínia
    -0.95
    Meli
    -0.95
    Ilustra
    -0.95
    AUSTR
    -0.93
    Hamb
    -0.93
     যে
    -0.92
    POSITIVE LOGITS
     here
    1.80
     included
    1.55
     we
    1.49
     today
    1.49
     instead
    1.42
     in
    1.37
     our
    1.34
     below
    1.20
     include
    1.17
     tetap
    1.16
    Activations Density 0.014%

    No Known Activations