© 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. 34-GEMMASCOPE-RES-131K
    7. 3025
    Prev
    Next
    INDEX
    Explanations

    some time / years

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

    The neuron detects expressions of duration emphasizing “for quite some time,” activating strongly on the words in that phrase.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     endnu
    -1.00
     olvidado
    -0.98
    たった
    -0.96
     dibujado
    -0.96
     måske
    -0.96
    nten
    -0.94
     cubierto
    -0.93
     forskellige
    -0.92
     Vgl
    -0.91
     cocción
    -0.90
    POSITIVE LOGITS
     some
    6.09
    Some
    4.09
    some
    3.84
     Some
    3.47
     algunos
    3.14
    一些
    3.13
     некоторых
    2.92
     некоторые
    2.91
     algunas
    2.83
     некото
    2.83
    Activations Density 0.051%

    No Known Activations