© 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. 27602
    Prev
    Next
    INDEX
    Explanations

    subject said or acted

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

    This neuron activates on occurrences of the third-person singular masculine pronoun “He” (especially at the start of clauses).

    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
     bereits
    -1.01
    ngOnInit
    -1.00
     manchmal
    -0.98
     уже
    -0.96
    cedor
    -0.94
    lieben
    -0.92
     поговори
    -0.91
     Ра
    -0.91
    最初
    -0.90
    Hvordan
    -0.89
    POSITIVE LOGITS
     also
    3.38
     furthermore
    2.03
     then
    1.98
     inoltre
    1.91
     further
    1.88
    also
    1.77
     também
    1.73
     также
    1.63
     said
    1.62
     tiež
    1.61
    Activations Density 0.046%

    No Known Activations