© 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. 35769
    Prev
    Next
    INDEX
    Explanations

    sports teams, Italian composer, versus other

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

    The neuron strongly activates on tokens that are parts of multiword compound terms (e.g., pick-up, steel-casing, enterprise resource planning).

    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
    -//
    -1.13
    ASHINGTON
    -1.09
    дравству
    -1.09
    opropyl
    -1.09
     niem
    -1.08
     tenha
    -1.08
     abuelo
    -1.08
    の詳細
    -1.08
     others
    -1.07
     linhas
    -1.05
    POSITIVE LOGITS
    Historique
    1.09
    Some
    1.04
    Commentaires
    1.03
    ⁉
    1.03
    いますが
    1.02
    它們
    1.02
     которые
    0.99
    Familien
    0.97
     brute
    0.95
    Parcours
    0.95
    Activations Density 0.009%

    No Known Activations