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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71209
    Prev
    Next
    INDEX
    Explanations

    troubling me a lot, has been driving me crazy all night

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

    The neuron fires on words and short phrases that signal strong emphasis or intensity (e.g. “a lot,” “crazy,” “nuts,” “so much,” “bothering me”).

    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
     iko
    -1.37
    кара
    -1.34
     rito
    -1.34
     milano
    -1.33
     sauf
    -1.32
     nocturna
    -1.31
     pinta
    -1.28
     pochette
    -1.25
     annet
    -1.22
     zijn
    -1.21
    POSITIVE LOGITS
     and
    1.92
     jamás
    1.45
     doua
    1.41
    พวก
    1.41
     podríamos
    1.38
    Superview
    1.37
    endroit
    1.34
     черво
    1.33
     éstas
    1.33
    が届
    1.32
    Activations Density 0.041%

    No Known Activations