© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 18083
    Prev
    Next
    INDEX
    Explanations

    string formatting width

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    ommer
    -0.07
    agama
    -0.07
     revol
    -0.07
     Revol
    -0.07
     poroz
    -0.07
     constituent
    -0.07
     constituents
    -0.07
     rel
    -0.06
    hari
    -0.06
    kje
    -0.06
    POSITIVE LOGITS
    æĭĻ
    0.05
    è¡¥é½IJ
    0.05
     vs
    0.05
    ill
    0.05
    min
    0.05
     Nash
    0.05
    èģĮå®Ī
    0.05
    ens
    0.05
    H
    0.05
    éĩıçļĦ
    0.04
    Activations Density 0.021%

    No Known Activations