© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130277
    Prev
    Next
    INDEX
    Explanations

    provided / known / intended / designed / developed

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

    The neuron selectively activates on past‐participle verb forms (especially in passive or technical descriptions, e.g. “provided,” “designed,” “installed,” “known”).

    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
     it
    -1.97
     Stunde
    -1.32
    おはよう
    -1.29
    ícias
    -1.27
     estos
    -1.23
     e
    -1.21
     they
    -1.20
     samodziel
    -1.19
     آن‌ها
    -1.19
    素晴らしい
    -1.18
    POSITIVE LOGITS
    蜮
    1.35
     särsk
    1.30
    𞤢
    1.29
     during
    1.23
     if
    1.21
    levance
    1.17
     gewiß
    1.15
     svensk
    1.15
     JsonResponse
    1.13
    んでいた
    1.12
    Activations Density 0.419%

    No Known Activations