© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12630
    Prev
    Next
    INDEX
    Explanations

    anger and frustration

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

    The neuron fires on words that convey strong emotional intensity or fervent opinion (e.g., anger, outrage, or intense reactions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    囁
    0.60
     আগ্রহী
    0.57
    腼
    0.57
    迷
    0.56
     रौनक
    0.55
    Yaml
    0.54
    心配
    0.54
    💒
    0.54
     Qualifications
    0.53
    ாலம்
    0.53
    POSITIVE LOGITS
     anger
    2.94
     angry
    2.67
     enraged
    2.61
     rage
    2.56
     angered
    2.45
     Anger
    2.42
     angrily
    2.41
    怒
    2.36
    愤怒
    2.33
     fury
    2.31
    Activations Density 0.468%

    No Known Activations