© 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. 1378
    Prev
    Next
    INDEX
    Explanations

    investigation or probe

    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
    大å¸Ī
    -0.09
    çĻ»
    -0.09
    zon
    -0.08
    å¾Ģè¿Ķ
    -0.08
    è¯Ī
    -0.08
     tắc
    -0.08
    许
    -0.08
    ses
    -0.08
    LECTION
    -0.08
     çĻ»
    -0.08
    POSITIVE LOGITS
     investigation
    0.16
    è°ĥæŁ¥
    0.15
     investigations
    0.15
     Investigations
    0.14
     Investigation
    0.13
    追究
    0.13
     investigate
    0.12
    è°ĥæŁ¥æĺ¾ç¤º
    0.12
     investigating
    0.12
    èª¿æŁ¥
    0.11
    Activations Density 0.086%

    No Known Activations