© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6746
    Prev
    Next
    INDEX
    Explanations

    unconventional or out of the ordinary

    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
    éĻ¶
    -0.10
    éľı
    -0.10
    çĤ«èĢĢ
    -0.10
    æĻĹ
    -0.10
    ä¼Ļ
    -0.09
    åĬ³ç´¯
    -0.08
    è§IJ
    -0.08
    çļĦèģĮä¸ļ
    -0.08
    è§Ĭ
    -0.08
    裥
    -0.08
    POSITIVE LOGITS
    çķ¸
    0.13
    åѤçĭ¬
    0.12
    æĭ§
    0.11
    åѤç«ĭ
    0.11
     lon
    0.11
    jected
    0.10
    amel
    0.10
    åĩ»
    0.09
    å¼±åĬ¿
    0.09
    ç»ıåİĨè¿ĩ
    0.09
    Activations Density 0.107%

    No Known Activations