© 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. 34-GEMMASCOPE-RES-131K
    7. 38568
    Prev
    Next
    INDEX
    Explanations

    indicates something is wrong

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

    The neuron fires on words signaling that something has “gone” or is “going” wrong (e.g. “it went wrong,” “going wrong,” “wrong,” “right?”).

    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
    からね
    -0.73
    yscy
    -0.72
    шки
    -0.71
    zja
    -0.71
    Bole
    -0.70
     enco
    -0.70
     Wahrheit
    -0.70
     gelo
    -0.70
    zembro
    -0.69
    Reserva
    -0.69
    POSITIVE LOGITS
     wrong
    2.81
    wrong
    1.49
     terribly
    1.48
     horribly
    1.30
     right
    1.27
    Wrong
    1.27
     fundamentally
    1.23
     Wrong
    1.19
     totally
    1.18
     WRONG
    1.16
    Activations Density 0.054%

    No Known Activations