© 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. 17834
    Prev
    Next
    INDEX
    Explanations

    estimates and estimations

    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
    èĿĩ
    -0.10
    ç½ij讯
    -0.10
    è´Ł
    -0.10
    è°ĭ
    -0.09
    rieve
    -0.09
    اط
    -0.09
    åij±
    -0.09
    pres
    -0.09
    æĥĨ
    -0.09
    è¹ĭ
    -0.09
    POSITIVE LOGITS
     estimate
    0.22
    估计
    0.20
     estimates
    0.19
    estimate
    0.19
     Estimates
    0.16
     Estimate
    0.16
    _estimate
    0.16
    ä¼°ç®Ĺ
    0.15
    ä¿Ŀå®Ī
    0.15
    认为
    0.12
    Activations Density 0.088%

    No Known Activations