© 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. 1280
    Prev
    Next
    INDEX
    Explanations

    asking for help

    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
    )))),
    -0.35
    -Cs
    -0.34
    vincial
    -0.30
     backpage
    -0.29
    UNET
    -0.29
     citiz
    -0.28
     Khá
    -0.27
    foundland
    -0.27
    pections
    -0.27
     ulus
    -0.27
    POSITIVE LOGITS
    éĿŀ常çļĦ
    0.28
     acest
    0.26
    好象
    0.25
     beside
    0.24
    æħ¢æħ¢çļĦ
    0.24
     physic
    0.24
    注æĺİåĩºå¤Ħ
    0.23
    éĥ¨ä»½
    0.23
    åħ¶å®ĥ
    0.23
     Aç
    0.23
    Activations Density 9.296%

    No Known Activations