© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91417
    Prev
    Next
    INDEX
    Explanations

    suspension or ban

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

    The neuron primarily detects numeric expressions—digits and numbers indicating quantities, durations, or counts.

    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
     detaine
    -0.90
     recoveries
    -0.82
     regimens
    -0.80
    瘘
    -0.77
     prosecutions
    -0.76
    纠
    -0.75
    ransition
    -0.75
    fontaine
    -0.72
    ienze
    -0.71
    ızı
    -0.71
    POSITIVE LOGITS
     suspension
    2.67
     suspensions
    2.02
     Suspension
    1.92
    suspension
    1.91
     suspended
    1.88
    Suspension
    1.85
     suspend
    1.51
     Suspended
    1.49
     SUSP
    1.47
     ban
    1.45
    Activations Density 0.021%

    No Known Activations