© 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. 57358
    Prev
    Next
    INDEX
    Explanations

    sorry followed by negation or past miss

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

    The neuron fires on emphatic, attention‐grabbing text—especially all-caps words and heavy punctuation (commas, exclamation marks, apostrophes) used for emphasis.

    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
     zvuky
    -1.48
     مختلف
    -1.42
    sprzedam
    -1.41
     problémy
    -1.40
     flotte
    -1.40
    meine
    -1.38
    vielen
    -1.36
    İyi
    -1.35
    jako
    -1.34
     növény
    -1.33
    POSITIVE LOGITS
     this
    1.91
     will
    1.54
     many
    1.52
     most
    1.49
     these
    1.49
     sometimes
    1.48
     similar
    1.48
     you
    1.45
     high
    1.41
     all
    1.38
    Activations Density 0.016%

    No Known Activations