© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15305
    Prev
    Next
    INDEX
    Explanations

    TEMPLATE GENERATED TESTCASE FILE

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

    The neuron detects boilerplate headers that mark code as auto-generated or template-generated (e.g. “This code was generated by…,” “TEMPLATE GENERATED TEST CASE FILE,” file/Art 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
    ところに
    -1.24
    ことができる
    -1.20
    ごとの
    -1.19
     człowiek
    -1.16
    なのかな
    -1.15
    の準備
    -1.11
     alemão
    -1.10
     acompañada
    -1.09
    のもの
    -1.09
    側から
    -1.09
    POSITIVE LOGITS
    では
    1.40
    を
    1.33
    があります
    1.23
     Grüßen
    1.22
    などを
    1.18
    は
    1.18
    <0xA3>
    1.15
    が
    1.12
    もあって
    1.10
    Masukkan
    1.09
    Activations Density 0.001%

    No Known Activations