© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 1795
    Prev
    Next
    INDEX
    Explanations

    specific identifiers

    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
    âĦħ
    -0.30
    ,eg
    -0.27
    âĪķ
    -0.27
    --------------↵
    -0.26
    [source
    -0.26
    ,ep
    -0.26
    âģĥ
    -0.25
    -----↵↵
    -0.25
     ()->
    -0.24
     CAUSED
    -0.24
    POSITIVE LOGITS
     burge
    0.34
     disg
    0.33
     sodom
    0.33
     lucr
    0.32
     ActionTypes
    0.31
     még
    0.30
     disadv
    0.29
     asiat
    0.29
    èIJľ
    0.28
    riger
    0.28
    Activations Density 17.973%

    No Known Activations