© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73929
    Prev
    Next
    INDEX
    Explanations

    urged or warned recipients

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

    The neuron detects verbs or phrases that issue calls to action or exhortations (e.g. urging, warning, calling on, appealing).

    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
    tranh
    -1.23
    utate
    -1.16
    ungkapkan
    -1.10
    وفير
    -1.10
    piscina
    -1.09
     desconhe
    -1.08
    entrevista
    -1.07
    uating
    -1.05
    认可
    -1.04
    puerta
    -1.04
    POSITIVE LOGITS
     all
    2.17
     everyone
    2.13
     anyone
    1.99
     you
    1.95
     everybody
    1.60
     young
    1.55
     them
    1.44
     anybody
    1.43
     any
    1.43
     people
    1.42
    Activations Density 0.013%

    No Known Activations