© 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. 1632
    Prev
    Next
    INDEX
    Explanations

    anti- words

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

    The neuron activates primarily on the prefix “anti-” appearing in compound words.

    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
    bathroom
    0.63
    ης
    0.60
     étage
    0.57
    ﻯ
    0.55
    ów
    0.55
    sheet
    0.54
     amiga
    0.53
     bathroom
    0.52
    cule
    0.52
    lijke
    0.52
    POSITIVE LOGITS
    словия
    0.44
    *((*
    0.42
     cannabinoids
    0.42
     fertilisers
    0.42
    คโนโล
    0.41
     государ
    0.41
    AsyncKeyState
    0.41
    とも
    0.40
     tides
    0.40
     zuf
    0.40
    Activations Density 0.006%

    No Known Activations