© 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. 73418
    Prev
    Next
    INDEX
    Explanations

    a/the/all followed by a specific term

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

    The neuron activates on numeric version identifiers (e.g. software or dependency version strings composed of digits, dots, and optional hyphen segments).

    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
    Ḍ
    -1.20
    wered
    -1.02
     fortæ
    -0.97
     adornos
    -0.94
     svart
    -0.94
     évé
    -0.94
     semillas
    -0.93
     nød
    -0.93
     joyas
    -0.93
     bordado
    -0.93
    POSITIVE LOGITS
     Эти
    1.09
     propagan
    1.02
     multiverse
    1.00
     hetzelfde
    0.99
     facili
    0.98
     dieselben
    0.98
     tarvit
    0.98
     combustibles
    0.98
     nephe
    0.97
     cardia
    0.97
    Activations Density 0.017%

    No Known Activations