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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100821
    Prev
    Next
    INDEX
    Explanations

    datelines locations

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

    This neuron activates on the sentence‐opening dash (–) that follows a location dateline (e.g., “CITY, State –”) in news articles.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     numerosos
    -1.65
     infuriating
    -1.62
     tervez
    -1.58
     verdaderas
    -1.52
     '
    -1.52
     ruptured
    -1.52
     maravillosas
    -1.51
     preciosas
    -1.49
    yordu
    -1.48
    ...'
    -1.45
    POSITIVE LOGITS
    .”
    2.13
     you
    1.76
     except
    1.70
     eds
    1.64
     única
    1.63
    并且
    1.61
     aren
    1.53
     is
    1.52
    }$.
    1.50
     '',
    1.50
    Activations Density 0.017%

    No Known Activations