© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 2688
    Prev
    Next
    INDEX
    Explanations

    only perfectly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
    羣æŃ£
    -0.25
    漪
    -0.23
    课åłĤæķĻåѦ
    -0.23
    æĻ®
    -0.22
    ä½ıå®ħ
    -0.22
    åĿ¤
    -0.21
    大èĥĨ
    -0.21
    .beta
    -0.21
     Prize
    -0.21
    æĹ©å°±
    -0.21
    POSITIVE LOGITS
     usual
    0.27
    usual
    0.26
    ä¾ĭè¡Į
    0.25
     mildly
    0.24
    -ish
    0.24
    æ¯Ķè¾ĥé«ĺ
    0.23
    normal
    0.22
    ish
    0.22
     sufficiently
    0.22
    秽
    0.22
    Activations Density 0.168%

    No Known Activations