© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 3007
    Prev
    Next
    INDEX
    Explanations

    This neuron isn’t looking for any semantic content at all but for where a token sits in the context window – it peaks on the earliest tokens and then steadily decays as you move further on in the sequence.

    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
    Shir
    0.88
    Kos
    0.88
     Tatiana
    0.86
    MOR
    0.81
    Carm
    0.77
    芴
    0.77
    Liber
    0.76
    Liz
    0.75
    Mor
    0.75
    Rita
    0.75
    POSITIVE LOGITS
     Pages
    0.80
    ʳ
    0.75
    WallArray
    0.73
     رغم
    0.73
     গণ্য
    0.71
     برانيه
    0.71
    рыв
    0.70
    чения
    0.69
    稆
    0.69
     اقسام
    0.68
    Activations Density 0.000%

    No Known Activations