© 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. 71516
    Prev
    Next
    INDEX
    Explanations

    descriptive nouns

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

    The neuron strongly activates on content words that serve as subjects or topics—particularly nouns (often domain-specific or technical nouns).

    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
     and
    -2.17
    にしました
    -1.38
     or
    -1.37
    にします
    -1.16
    nimmt
    -1.15
    してくれた
    -1.09
    gaande
    -1.09
    หรือ
    -1.09
     tocado
    -1.09
     БО
    -1.09
    POSITIVE LOGITS
     is
    1.67
     are
    1.58
     jeudi
    1.11
     คือ
    1.05
     vendredi
    1.02
     mercredi
    1.02
     valore
    1.01
     always
    0.98
    ですが
    0.97
    is
    0.97
    Activations Density 0.070%

    No Known Activations