© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 17859
    Prev
    Next
    INDEX
    Explanations

    result or consequence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    oo
    -0.09
    çļĦéĩįçĤ¹
    -0.08
    å°±è§īå¾Ĺ
    -0.08
     reluct
    -0.08
    oom
    -0.08
    oop
    -0.08
    çŃĿ
    -0.08
    oot
    -0.08
    ihad
    -0.08
    å®´
    -0.08
    POSITIVE LOGITS
    çļĦç»ĵæŀľ
    0.16
     ÑĢезÑĥлÑĮÑĤаÑĤе
    0.15
     consequence
    0.15
     result
    0.14
    åIJİæŀľ
    0.13
    æºIJäºİ
    0.13
    (result
    0.12
    .result
    0.12
     consequences
    0.12
     resultado
    0.12
    Activations Density 0.081%

    No Known Activations