© 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. 27820
    Prev
    Next
    INDEX
    Explanations

    personal thoughts and feelings

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

    The neuron strongly activates on first-person expressions of desire, opinion, or feeling (e.g. “I want,” “I wish,” “I think,” “I feel,” “I like”).

    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
     anuncios
    -1.14
     usamos
    -1.10
     provecho
    -1.02
     repuesto
    -1.02
     assortment
    -1.02
    Poloha
    -1.02
     conocemos
    -1.01
     evidence
    -1.00
    <bos>
    -1.00
     pieces
    -1.00
    POSITIVE LOGITS
     to
    1.59
     that
    1.47
     this
    1.38
     from
    1.16
     with
    1.09
    了
    1.02
     since
    0.99
    imagination
    0.99
     when
    0.99
     will
    0.99
    Activations Density 0.081%

    No Known Activations