© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 856
    Prev
    Next
    INDEX
    Explanations

    technical description

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bureaucr
    -0.36
     acknow
    -0.28
     fick
    -0.27
    .TODO
    -0.27
     disadv
    -0.27
     enthus
    -0.27
    rible
    -0.27
    éparation
    -0.25
    çļĦ带é¢Ĩ
    -0.25
     beneficiation
    -0.25
    POSITIVE LOGITS
    æĸ¹æīį
    0.32
    å·²çĦ¶
    0.31
    ﹩
    0.30
    便åı¯
    0.27
    åħĪåīį
    0.27
     веÑģÑĮма
    0.26
    æĩĴå¾Ĺ
    0.25
    诸å¦Ĥ
    0.25
    åĿĩå·²
    0.24
    åĬ¡å¿ħ
    0.24
    Activations Density 12.405%

    No Known Activations