© 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. 22-GEMMASCOPE-RES-131K
    7. 14595
    Prev
    Next
    INDEX
    Explanations

    explaining what something is

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

    This neuron activates on the main subject or topic noun (or pronoun) of a clause—that is, the key “who” or “what” the sentence is about.

    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
     csak
    -1.13
     только
    -1.01
     only
    -0.99
    eté
    -0.97
     hanya
    -0.94
     sadece
    -0.92
     moglie
    -0.91
    șa
    -0.91
    Rechts
    -0.90
     YANG
    -0.88
    POSITIVE LOGITS
     about
    1.63
     amounted
    1.16
     mean
    1.13
     denomina
    0.95
     doing
    0.93
     aconte
    0.90
     entails
    0.90
    Ⓣ
    0.89
     ;
    0.88
     shape
    0.88
    Activations Density 0.091%

    No Known Activations