© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 68163
    Prev
    Next
    INDEX
    Explanations

    lack of action or failure

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

    The neuron fires on strong accusatory or negative-judgment words (e.g. “silent,” “complicit,” “failing,” “not,” “enforcement”) that signal criticism or blame.

    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
     tertentu
    -1.13
     Fakta
    -1.02
     sebag
    -1.00
     sebaik
    -0.98
     pira
    -0.98
     mami
    -0.97
     raya
    -0.97
     strass
    -0.95
     traktor
    -0.94
     papi
    -0.93
    POSITIVE LOGITS
     inaction
    1.48
     action
    1.27
     response
    1.21
     responds
    1.04
     support
    1.02
    ksjon
    1.02
    nada
    0.99
    responsibility
    0.99
     eficaz
    0.98
     รับ
    0.98
    Activations Density 0.029%

    No Known Activations