© 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. 20045
    Prev
    Next
    INDEX
    Explanations

    upset or distraught emotions

    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
    ï¼Ĭ
    -0.06
    æĬĬ
    -0.06
    zin
    -0.06
    nist
    -0.06
    saus
    -0.06
    uth
    -0.06
    zit
    -0.05
    nk
    -0.05
    kins
    -0.05
    schlä
    -0.05
    POSITIVE LOGITS
     upset
    0.08
    enda
    0.06
     upsetting
    0.06
    argent
    0.06
    enje
    0.06
    گرد
    0.06
    ponder
    0.06
    amento
    0.05
    ÙĬÙĤ
    0.05
    ov
    0.05
    Activations Density 0.006%

    No Known Activations