© 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. 34574
    Prev
    Next
    INDEX
    Explanations

    Guggenheim, can, copyright

    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
    åIJĦæł·
    -0.10
    çĽĨåľ°
    -0.09
    æ¤İ
    -0.09
    enson
    -0.09
    çľ¼çķĮ
    -0.09
    .communication
    -0.09
    elter
    -0.09
    _NUMERIC
    -0.09
    ered
    -0.09
    æ²ŁéĢļ
    -0.09
    POSITIVE LOGITS
    6
    0.11
    5
    0.11
    Ñĩа
    0.11
    4
    0.10
    1
    0.10
    7
    0.10
    8
    0.10
    3
    0.10
    2
    0.09
    9
    0.09
    Activations Density 0.028%

    No Known Activations