© 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. 55687
    Prev
    Next
    INDEX
    Explanations

    mesosphere.marathon

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

    The neuron strongly activates on tokens containing the substring “mes” (e.g. in “mesosphere,” “mesos,” “mesangial,” “Mesocestoides,” 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_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
    生长
    -0.73
     CORS
    -0.72
    maco
    -0.70
    douard
    -0.70
    ั่น
    -0.69
    答案
    -0.69
     貝
    -0.68
    Technically
    -0.68
    brador
    -0.67
    ebenarnya
    -0.67
    POSITIVE LOGITS
    いで
    0.84
     Mes
    0.84
    лете
    0.75
     meson
    0.74
    崩
    0.73
    ях
    0.71
    ("#
    0.71
    impulse
    0.70
     téléchargement
    0.70
     Compromise
    0.69
    Activations Density 0.026%

    No Known Activations