© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 10923
    Prev
    Next
    INDEX
    Explanations

    system include headers

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

    system-level C header file includes, particularly those in the "sys/" directory.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires strongly on tokens from POSIX/system include directives (e.g. “unistd,” “sys/types,” “sys/wait,” etc.), so it is detecting system‐level header includes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    隳
    -1.62
    很赞
    -1.53
     vítima
    -1.45
    嘞
    -1.42
    一旁的
    -1.41
     amea
    -1.41
     called
    -1.39
     które
    -1.36
    很喜欢
    -1.34
    英語版
    -1.33
    POSITIVE LOGITS
    \
    1.41
    涣
    1.33
    ธุ
    1.32
    }
    1.30
    𝑮
    1.28
    爱情
    1.26
     .
    1.25
    diente
    1.24
    IC
    1.21
     etmek
    1.21
    Activations Density 0.022%

    No Known Activations