© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 20731
    Prev
    Next
    INDEX
    Explanations

    what the

    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
    ry
    -0.06
    ability
    -0.05
    व
    -0.05
    ħ
    -0.05
    åĽ½å®¶æ£®æŀĹåħ¬åĽŃ
    -0.05
    ÙĪÙī
    -0.05
    éĤ£ç§į
    -0.05
    to
    -0.05
    ro
    -0.05
    ru
    -0.05
    POSITIVE LOGITS
    hoven
    0.07
    aad
    0.07
    è´¨çļĦ
    0.06
    inges
    0.06
    ahr
    0.06
    atest
    0.06
    amedi
    0.06
     اÙĦرسÙħÙĬ
    0.06
    etros
    0.06
    geler
    0.06
    Activations Density 0.012%

    No Known Activations