© 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. 22-GEMMASCOPE-RES-131K
    7. 26060
    Prev
    Next
    INDEX
    Explanations

    message components

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

    This neuron activates on the determiners or introductory words (like “a,” “the,” “first,” “his”) that begin each item in a bulleted or enumerated list.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     téléchargement
    -0.98
     manuelle
    -0.93
     kritis
    -0.90
     omfatt
    -0.90
    MMMM
    -0.88
     emberek
    -0.88
     kampen
    -0.87
    recently
    -0.87
    lē
    -0.85
     віта
    -0.84
    POSITIVE LOGITS
     then
    1.18
     later
    1.07
    AllAfrica
    0.91
     perennial
    0.85
     vervolgens
    0.84
     hanggang
    0.84
     unquestion
    0.84
     sound
    0.82
     customary
    0.81
    deco
    0.81
    Activations Density 0.042%

    No Known Activations