© 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. 24-GEMMASCOPE-2-RES-16K
    7. 1623
    Prev
    Next
    INDEX
    Explanations

    if you

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

    This neuron activates on the word “If,” flagging conditional clauses that introduce “if … then” statements.

    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_24_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
     কিনা
    0.68
     минуту
    0.66
     Idiot
    0.66
     reyes
    0.63
     నేను
    0.62
    았던
    0.61
    했
    0.61
    갔
    0.61
    ている
    0.61
    았
    0.61
    POSITIVE LOGITS
    framework
    0.66
    scan
    0.59
    },
    0.55
    ৫
    0.55
    та
    0.54
    scale
    0.53
     framework
    0.53
    }$,
    0.52
     spatial
    0.52
    prehensive
    0.52
    Activations Density 0.001%

    No Known Activations