© 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. 10-GEMMASCOPE-RES-131K
    7. 40487
    Prev
    Next
    INDEX
    Explanations

    past perfect actions/states

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

    The neuron primarily detects simple‐tense verbs (base‐form and past‐tense/past‐participle verbs) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     REALLY
    -1.95
     VERY
    -1.70
     unbelievably
    -1.59
     somente
    -1.55
     seleccionados
    -1.53
     insanely
    -1.51
     amazingly
    -1.50
     فقط
    -1.50
     некоторых
    -1.45
     myriad
    -1.44
    POSITIVE LOGITS
     during
    2.11
     at
    2.09
     in
    2.02
     any
    2.00
     this
    1.99
     allowing
    1.84
     around
    1.83
     before
    1.73
     while
    1.69
     regardless
    1.69
    Activations Density 0.045%

    No Known Activations