© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 14492
    Prev
    Next
    INDEX
    Explanations

    how I feel

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

    The neuron fires on verbs and phrases that signal expressing or stating one’s feelings or opinions (e.g. “tell you how,” “can’t express how,” “state how,” “begin to thank”).

    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
     it
    -1.41
    ↵↵
    -1.34
     What
    -1.27
     isn
    -1.27
     such
    -1.26
     does
    -1.24
     one
    -1.23
     here
    -1.20
     doesn
    -1.18
     any
    -1.16
    POSITIVE LOGITS
     of
    1.78
     innymi
    1.50
    珦
    1.45
     другим
    1.42
     my
    1.41
    他の
    1.39
     otras
    1.37
     другими
    1.37
     innych
    1.34
    ㄘ
    1.34
    Activations Density 0.017%

    No Known Activations