© 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. 5833
    Prev
    Next
    INDEX
    Explanations

    make an impact

    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
    å²Ķ
    -0.09
    cel
    -0.08
    å¼±
    -0.08
    å¼±çĤ¹
    -0.08
    Straight
    -0.08
    cla
    -0.08
    第ä¸īèĢħ
    -0.08
    éĢ¶
    -0.08
    oss
    -0.08
    erk
    -0.08
    POSITIVE LOGITS
     impact
    0.19
    çļĦå½±åĵį
    0.18
    å½±åĵį
    0.18
    å½±åĵįåĬĽ
    0.17
    impact
    0.17
    æ·±è¿ľ
    0.16
    å½±éŁ¿
    0.15
     влиÑı
    0.15
     Impact
    0.15
    Impact
    0.14
    Activations Density 0.054%

    No Known Activations