© 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. 15596
    Prev
    Next
    INDEX
    Explanations

    teacher

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

    The neuron fires on words related to teaching—especially the token “teacher” (and its variants like “teachers” or “teaching”).

    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
     \
    -2.38
    ’
    -2.36
     –
    -2.33
    的一天
    -2.28
    ”
    -2.27
     /
    -2.19
    '
    -2.17
     '
    -2.14
    ,'
    -2.13
     drenched
    -2.11
    POSITIVE LOGITS
    .
    2.33
    蔸
    2.22
     メニュー
    2.20
    狆
    2.17
    </td>
    2.16
    😨
    2.14
    teacher
    2.13
    觔
    2.09
     发型
    2.08
     Saltar
    2.03
    Activations Density 0.005%

    No Known Activations