© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 13440
    Prev
    Next
    INDEX
    Explanations

    comment section

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

    The neuron is triggered by occurrences of the word “comment” (e.g. in code/documentation comment markers or the word “comment” itself).

    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
    Více
    -1.16
     you
    -1.07
     inflatable
    -1.05
     pool
    -1.04
    grats
    -1.03
     of
    -1.02
    Uninitialized
    -1.02
     Reised
    -1.01
    plings
    -1.01
    restler
    -1.00
    POSITIVE LOGITS
     sambung
    1.16
     about
    1.16
    鏝
    1.06
     there
    1.01
     cualquier
    1.01
     componenti
    1.00
    usion
    0.99
     its
    0.97
     canzone
    0.96
     strade
    0.95
    Activations Density 0.040%

    No Known Activations