© 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. 1522
    Prev
    Next
    INDEX
    Explanations

    Breaking or negative events

    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
     öde
    -0.30
    ;;;;;;;;
    -0.26
    ÑĢоÑģл
    -0.23
    å¦Ĥæŀľçݩ家
    -0.23
    "w
    -0.23
    }elseif
    -0.23
    Ø£ÙĪض
    -0.22
    åĩºè½¨
    -0.22
    æį¢åı¥è¯Ŀ
    -0.22
    .Formatter
    -0.22
    POSITIVE LOGITS
    busters
    0.29
    obierno
    0.28
    ĵĺ
    0.28
    portun
    0.25
     BusinessException
    0.25
    lashes
    0.25
    úb
    0.24
    traî
    0.24
    aneously
    0.23
    variably
    0.23
    Activations Density 6.694%

    No Known Activations