© 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. 30343
    Prev
    Next
    INDEX
    Explanations

    ignored and denied

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

    The neuron responds strongly to very high‐frequency “function” tokens (common stop-words like to, is, no, or, which, as, been, etc., and punctuation such as commas).

    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
    膵
    -0.84
    Simulator
    -0.83
    ставля
    -0.82
    вяза
    -0.82
     afecta
    -0.80
     Solu
    -0.80
     Simulator
    -0.79
     dégust
    -0.78
     غذ
    -0.77
     punishment
    -0.76
    POSITIVE LOGITS
     ignore
    1.25
    Ignore
    1.23
     tramp
    1.22
     obliter
    1.16
     nul
    1.12
     disregard
    1.09
     ignores
    1.07
    eclipse
    1.07
     sacrificed
    1.04
     neglect
    1.03
    Activations Density 0.041%

    No Known Activations