© 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. 10-GEMMASCOPE-RES-131K
    7. 128585
    Prev
    Next
    INDEX
    Explanations

    Samuel followed by names

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

    The neuron fires on mentions of the proper name “Samuel.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     '
    -3.59
     "
    -3.19
    ...."
    -2.45
     endet
    -2.42
    !"
    -2.39
     poderosos
    -2.38
     schmeckt
    -2.34
    。"
    -2.33
     -
    -2.27
     periódicos
    -2.27
    POSITIVE LOGITS
    “.
    3.64
    .“
    3.55
    —“
    3.42
    ”،
    3.22
    The
    3.09
    ”.
    2.98
    </td>
    2.97
    -“
    2.95
    ”…
    2.92
    (“
    2.86
    Activations Density 0.002%

    No Known Activations