© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14830
    Prev
    Next
    INDEX
    Explanations

    why planet explanation

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

    This neuron flags rare or low-frequency subword tokens (uncommon words or word parts).

    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
    ни
    0.77
    ли
    0.77
    р
    0.71
    樣
    0.71
    ви
    0.70
    a
    0.70
    য়
    0.67
     доку
    0.67
    ه
    0.67
    ר
    0.67
    POSITIVE LOGITS
    ﻢ
    0.80
    ు
    0.79
     तले
    0.74
     Bedingungen
    0.72
    髡
    0.72
    뜹
    0.70
     thunderstorm
    0.70
    ረት
    0.69
     policeman
    0.68
     blijkt
    0.68
    Activations Density 0.001%

    No Known Activations