© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 86752
    Prev
    Next
    INDEX
    Explanations

    accepting risk

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

    The neuron is primarily triggered by numerical expressions—digits, years, counts, and other number‐related tokens.

    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
    viron
    -0.91
    ouchers
    -0.81
    omal
    -0.79
    enoord
    -0.75
    xpress
    -0.75
     korun
    -0.75
     bateaux
    -0.75
    ownload
    -0.74
     نے
    -0.74
    FactoryBean
    -0.73
    POSITIVE LOGITS
     risk
    6.44
    risk
    4.94
     Risk
    4.81
    Risk
    4.75
     risks
    4.75
     risked
    4.66
     risking
    4.41
     RISK
    4.13
     riesgo
    4.03
    RISK
    3.64
    Activations Density 0.077%

    No Known Activations