© 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. 26770
    Prev
    Next
    INDEX
    Explanations

    absolute value or basis

    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
    éĹ°
    -0.09
    ä¿¡èªī
    -0.09
    åĩ¡äºĭ
    -0.08
    举åįĹ
    -0.08
    ECT
    -0.08
     Lad
    -0.08
    æľĹ
    -0.08
    abl
    -0.08
     Estimates
    -0.07
    æ¹Ľ
    -0.07
    POSITIVE LOGITS
     absolute
    0.17
    ç»Ŀ对
    0.16
    absolute
    0.16
    _absolute
    0.15
    Absolute
    0.15
     Absolute
    0.15
     абÑģолÑİÑĤ
    0.14
    Relative
    0.13
     relative
    0.13
    .absolute
    0.13
    Activations Density 0.130%

    No Known Activations