© 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. 13086
    Prev
    Next
    INDEX
    Explanations

    every followed by fear or second followed by Sal

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

    The neuron strongly activates on decimal or floating‐point numeric tokens.

    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
    ↵
    0.38
     έως
    0.33
    racing
    0.33
     ศ
    0.32
     Chaud
    0.31
    solutions
    0.31
     を
    0.30
     मैचों
    0.30
     Mér
    0.30
    Solutions
    0.30
    POSITIVE LOGITS
     _____
    0.38
    🫡
    0.38
    ⚠️
    0.37
     avantages
    0.36
    కోవడం
    0.36
    ⏤
    0.36
    這樣子
    0.36
    🟣
    0.35
    ﻒ
    0.35
     проє
    0.35
    Activations Density 1.345%

    No Known Activations