© 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. 2038
    Prev
    Next
    INDEX
    Explanations

    also

    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
     either
    -0.28
    variably
    -0.28
     actually
    -0.27
    ä¸Ģæĺ¯
    -0.27
    å°¤åħ¶æĺ¯
    -0.27
    ä¸įä½Ĩ
    -0.27
    either
    -0.26
    ä¹ĥèĩ³
    -0.25
    çĶļèĩ³
    -0.25
    å°¤åħ¶
    -0.25
    POSITIVE LOGITS
    verständlich
    0.26
    -scalable
    0.26
    ^K
    0.24
     démarch
    0.24
    astle
    0.24
    chers
    0.24
     ë¶Ī구íķĺê³ł
    0.24
    ("(%
    0.23
    swire
    0.23
    å°±æĿ¥çľĭçľĭ
    0.23
    Activations Density 2.415%

    No Known Activations