© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 7415
    Prev
    Next
    INDEX
    Explanations

    words forming descriptive adjectives and adverbs

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

    The neuron strongly activates on unusually long, rare multisyllabic words (especially emphatic adjectives, adverbs, or verbs).

    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
     countless
    -1.50
     any
    -1.48
     other
    -1.47
     many
    -1.44
     innumerable
    -1.38
     same
    -1.36
     only
    -1.35
     numerous
    -1.27
    これからも
    -1.25
     can
    -1.23
    POSITIVE LOGITS
     wła
    1.34
    TestingModule
    1.34
    ,
    1.33
    向量
    1.30
    ñar
    1.29
    ❓
    1.28
    untuk
    1.28
    sonar
    1.25
    、
    1.25
    ?」
    1.24
    Activations Density 0.389%

    No Known Activations