© 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. 38563
    Prev
    Next
    INDEX
    Explanations

    news articles

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

    The neuron strongly activates on apology and remorse expressions (e.g., “sorry,” “please forgive us,” “we’re really sorry,” 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
     I
    -1.04
     its
    -0.94
     but
    -0.90
     due
    -0.86
     Wirksamkeit
    -0.84
    ليج
    -0.82
    ",
    -0.82
    clause
    -0.81
    IntentFilter
    -0.80
    즉
    -0.80
    POSITIVE LOGITS
     теперь
    0.92
    Exc
    0.91
     nasıl
    0.88
    Ress
    0.87
    gemeinschaft
    0.86
     ahora
    0.85
     natuurlijk
    0.85
     zodat
    0.84
     henceforth
    0.84
    glicher
    0.83
    Activations Density 0.058%

    No Known Activations