© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70764
    Prev
    Next
    INDEX
    Explanations

    choosing or searching for

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

    The neuron flags instructional/advisory phrasing—especially modal verbs and guidance cues (may, ought, must, should, first, how to, 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_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
    ícias
    -1.26
    gering
    -1.25
    erdings
    -1.16
     we
    -1.16
     situe
    -1.15
    atine
    -1.13
    ἓ
    -1.10
    軋
    -1.06
    licante
    -1.03
    atonic
    -1.03
    POSITIVE LOGITS
     your
    1.55
     who
    1.41
    ֶּ
    1.32
     services
    1.30
     trust
    1.27
    xiaomi
    1.27
    ַּ
    1.26
     dépenses
    1.26
    匿名
    1.23
    ึกษา
    1.23
    Activations Density 0.057%

    No Known Activations