© 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. 28258
    Prev
    Next
    INDEX
    Explanations

    logged in status

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

    The main thing this neuron does is detect mentions of logging in or being logged in (various forms of “log in,” “logged,” “logging,” etc.).

    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
    ↵↵
    -2.81
     {
    -2.28
    缡
    -2.17
    v
    -2.06
     Which
    -2.05
     was
    -1.86
     let
    -1.84
    g
    -1.81
    และ
    -1.77
     ซึ่ง
    -1.77
    POSITIVE LOGITS
    <bos>
    2.38
    睒
    2.20
    日本的
    2.09
    ⃚
    2.08
    Only
    2.06
    ."));
    2.05
    𞤢
    2.02
     veranstal
    2.02
    Depending
    1.98
     verſ
    1.97
    Activations Density 0.005%

    No Known Activations