© 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. 22-GEMMASCOPE-RES-131K
    7. 35877
    Prev
    Next
    INDEX
    Explanations

    before or original state

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

    The neuron fires on informal expressive punctuation stretches—e.g. long runs of dots (“…”, “..”) or exclamation points—and similar chat‐style intensifiers in casual text.

    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
    Unfortunately
    -1.01
     Unfortunately
    -0.99
     unfortunately
    -0.87
     Regret
    -0.87
     leider
    -0.86
     ويكيپيديا
    -0.86
     уже
    -0.84
     tecnológico
    -0.83
     Sadly
    -0.82
     già
    -0.79
    POSITIVE LOGITS
     initially
    1.63
     earlier
    1.60
     previously
    1.38
     previous
    1.38
     initial
    1.26
    earlier
    1.25
    previously
    1.24
     anfangs
    1.21
     sempat
    1.20
     originally
    1.18
    Activations Density 0.082%

    No Known Activations