© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65703
    Prev
    Next
    INDEX
    Explanations

    hasBeen

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

    The neuron fires on past-participle and perfect-tense verb forms (especially “is/has been/been/used/maintained/serviced,” etc.) that describe an item’s condition or history.

    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
    momentum
    -0.93
    titions
    -0.92
     deň
    -0.91
    ΙΝ
    -0.88
     cheaply
    -0.87
     天然
    -0.86
    шому
    -0.86
     iniciales
    -0.84
    zama
    -0.84
     barato
    -0.84
    POSITIVE LOGITS
     بوده
    1.17
     prior
    1.16
     since
    1.14
    never
    1.12
    тивным
    1.10
    antenimiento
    1.05
     restoran
    1.05
     médec
    1.02
    ложений
    1.02
     last
    1.00
    Activations Density 0.016%

    No Known Activations