© 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. 34-GEMMASCOPE-RES-131K
    7. 66238
    Prev
    Next
    INDEX
    Explanations

    Kin selection, Field of, Reduced gradient, e prime

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

    The neuron strongly activates on the very first token of each new text segment (i.e. the opening word or number at the start of a block).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     éter
    -0.99
     septembre
    -0.98
     verden
    -0.97
     arkite
    -0.94
     consulté
    -0.94
    🏫
    -0.93
     chrétien
    -0.90
    rugu
    -0.88
     salg
    -0.87
    vaných
    -0.87
    POSITIVE LOGITS
     molte
    0.93
     extremely
    0.92
    9
    0.83
     festen
    0.80
     fenti
    0.79
    τε
    0.79
    jsonp
    0.79
     spectacular
    0.78
    ier
    0.78
     alimenti
    0.77
    Activations Density 0.000%

    No Known Activations