© 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. 21055
    Prev
    Next
    INDEX
    Explanations

    end of sentence punctuation

    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
    çĥŃ线
    -0.06
    ént
    -0.06
    ocas
    -0.06
    zana
    -0.06
    ichtlich
    -0.06
    æĹłå¸¸
    -0.06
    akes
    -0.06
    киÑı
    -0.06
    åħ¶
    -0.06
    _typeof
    -0.06
    POSITIVE LOGITS
    arkin
    0.06
     wont
    0.05
     verr
    0.05
    个æĸ¹éĿ¢
    0.05
    quote
    0.05
     mult
    0.05
    fty
    0.05
     beh
    0.05
    jab
    0.05
    abung
    0.05
    Activations Density 0.011%

    No Known Activations