© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49441
    Prev
    Next
    INDEX
    Explanations

    academic subjects and teaching

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

    This neuron fires strongly on words that appear as section‐heading labels or category titles in educational materials (e.g. “Benefits,” “Academic,” “Subject,” “Classroom,” “Teaching,” 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
    álu
    -0.85
     STUDENT
    -0.81
    ônico
    -0.75
     poro
    -0.75
     Learners
    -0.73
    Якщо
    -0.73
     prece
    -0.72
    hôpital
    -0.71
     anhyd
    -0.71
     chup
    -0.70
    POSITIVE LOGITS
     subjects
    5.06
     subject
    4.88
    subject
    4.06
    subjects
    3.97
     Subjects
    3.75
    Subject
    3.66
    Subjects
    3.59
     Subject
    3.36
    SUBJECT
    2.98
     SUBJECTS
    2.92
    Activations Density 0.036%

    No Known Activations