© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66356
    Prev
    Next
    INDEX
    Explanations

    start of sentences and code

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

    The neuron strongly activates on tokens written entirely in uppercase (e.g. acronyms, metadata markers, or rush‐transcript labels).

    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
     cukru
    -1.40
     celebridades
    -1.14
     この
    -1.09
     повіт
    -1.02
    arakhand
    -1.02
     filós
    -1.00
     كل
    -0.99
     my
    -0.98
     honlap
    -0.96
     I
    -0.96
    POSITIVE LOGITS
    本日は
    1.12
    💒
    1.05
     gammal
    1.05
    ματος
    1.02
    icular
    0.99
     paladin
    0.99
     جاده
    0.98
     Mojave
    0.98
     rupiah
    0.97
     古い
    0.97
    Activations Density 0.000%

    No Known Activations