© 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. 18532
    Prev
    Next
    INDEX
    Explanations

    multiple different various many several

    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
    LOCKS
    -0.10
    èĥĻ
    -0.10
    人æŃ»äº¡
    -0.10
    éĺ¶
    -0.10
    วà¸ĩศ
    -0.09
    _THAN
    -0.09
    asti
    -0.09
    placement
    -0.09
    éªĪ
    -0.09
    ulo
    -0.09
    POSITIVE LOGITS
    å¤ļç§į
    0.14
     different
    0.13
     multiple
    0.13
    ä¸įåIJĮçļĦ
    0.13
     together
    0.12
     elements
    0.12
    å¤ļ个
    0.12
    å¤ļ家
    0.12
    为ä¸Ģä½ĵ
    0.12
    äºİä¸Ģä½ĵ
    0.11
    Activations Density 0.186%

    No Known Activations