© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70855
    Prev
    Next
    INDEX
    Explanations

    invalid or undefined results

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

    The neuron fires on terms that signal error, invalidity, or nonphysical/undefined conditions (e.g. singularities, invalid formats, undefined behavior, negative‐weight edges).

    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
     chocolat
    -1.07
     lópez
    -1.05
     occuper
    -1.02
    циали
    -1.01
    CESO
    -1.00
     搞笑
    -0.99
    儇
    -0.99
     ваз
    -0.96
    睁开
    -0.96
    と同
    -0.96
    POSITIVE LOGITS
     when
    1.13
     because
    1.08
     invalid
    1.02
     ciertas
    0.98
    قرير
    0.96
     whenever
    0.96
    كوين
    0.94
     Whenever
    0.93
     produced
    0.91
     resulted
    0.90
    Activations Density 0.063%

    No Known Activations