© 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. 1996
    Prev
    Next
    INDEX
    Explanations

    comparative and dependent descriptions

    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
    еÑģÑĮ
    -0.24
     forever
    -0.23
    gem
    -0.22
    åŃĺåľ¨äºİ
    -0.22
    [
    -0.22
     ê°ģì¢ħ
    -0.21
     Trick
    -0.21
     organizing
    -0.21
    æĺ¯ä¸Ģ个
    -0.21
     Mines
    -0.21
    POSITIVE LOGITS
     compared
    0.36
     consistently
    0.33
    æŀľçĦ¶
    0.32
    ç»ĵæŀľæĺ¾ç¤º
    0.31
     Compared
    0.30
     decreased
    0.29
     surprisingly
    0.29
     correctly
    0.28
     despite
    0.28
    缸æ¯Ķäºİ
    0.28
    Activations Density 0.626%

    No Known Activations