© 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. 27541
    Prev
    Next
    INDEX
    Explanations

    assembly code registers

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

    This neuron detects occurrences of CPU register names (e.g. “%ecx,” “rax,” “rdi,” etc.) in assembly code.

    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
     our
    -1.76
     your
    -1.41
     really
    -1.35
    孔子
    -1.34
     triliun
    -1.30
     naprawdę
    -1.29
     their
    -1.23
    になったら
    -1.20
     ovat
    -1.20
     all
    -1.20
    POSITIVE LOGITS
    などの
    1.32
     チラ
    1.22
    见的
    1.18
    がありますが
    1.17
     corrid
    1.17
     cerim
    1.16
     innumer
    1.16
     byg
    1.13
    使い方
    1.12
    你能
    1.12
    Activations Density 0.004%

    No Known Activations