© 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. 2430
    Prev
    Next
    INDEX
    Explanations

    fix them

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

    This neuron activates on verbs that denote repairing, fixing, or otherwise remedying and improving something.

    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
     jugo
    0.60
     जिक्र
    0.59
    ρί
    0.58
    ц
    0.57
     precautionary
    0.57
    炬
    0.57
    鵡
    0.56
    ск
    0.55
    भो
    0.54
    ุ่น
    0.52
    POSITIVE LOGITS
     Repair
    1.22
    Repair
    1.21
     repair
    1.15
    修复
    1.15
     napraw
    1.15
     repairing
    1.14
     restore
    1.12
     recover
    1.09
     reparar
    1.09
     repaired
    1.09
    Activations Density 0.345%

    No Known Activations