© 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. 21925
    Prev
    Next
    INDEX
    Explanations

    crossing or reaching a line

    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
    ç¨Ĺ
    -0.11
    æĪ´
    -0.11
    æIJº
    -0.11
    ç¬ı
    -0.10
    FRING
    -0.10
    åij·
    -0.10
    oron
    -0.10
    为空
    -0.10
    ä¸Ģéĥ¨
    -0.09
    éĨ®
    -0.09
    POSITIVE LOGITS
     line
    0.15
    line
    0.13
    çķĮéĻIJ
    0.13
    线
    0.12
    éĺĪ
    0.11
    LINE
    0.11
    红线
    0.11
     Line
    0.11
    Line
    0.11
    -line
    0.10
    Activations Density 0.047%

    No Known Activations