© 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. 32978
    Prev
    Next
    INDEX
    Explanations

    restrictions and failures

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

    The neuron fires on terms that signal negative outcomes or failures (e.g. exclusions, breaches, losses, divergences, unsubstantiated events, etc.).

    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
     только
    -1.39
     only
    -1.16
     nhưng
    -1.13
     but
    -1.13
     però
    -1.04
     แต่
    -0.97
     ancak
    -0.96
    僅
    -0.95
     však
    -0.95
     sparingly
    -0.94
    POSITIVE LOGITS
     every
    1.30
     even
    1.30
     everywhere
    1.14
     most
    1.10
    inerario
    1.07
     all
    1.03
     because
    0.98
     både
    0.96
     unless
    0.94
    >>>>>>>>
    0.94
    Activations Density 0.087%

    No Known Activations