© 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. 17684
    Prev
    Next
    INDEX
    Explanations

    coming to terms with

    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
    uggage
    -0.05
     =>$
    -0.05
    ul
    -0.05
    çıŃ级管çIJĨ
    -0.05
     wind
    -0.05
     hod
    -0.05
     ch
    -0.05
    風ãģ«
    -0.05
    él
    -0.05
    稱çĤº
    -0.05
    POSITIVE LOGITS
    illis
    0.06
     ноÑĢмалÑĮно
    0.06
    æŃ£å¸¸
    0.06
    umah
    0.06
    çļĦæŃ£å¸¸
    0.06
    oscop
    0.06
    aiss
    0.05
    :disable
    0.05
    vorg
    0.05
     normally
    0.05
    Activations Density 0.008%

    No Known Activations