© 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. 26501
    Prev
    Next
    INDEX
    Explanations

    offense or defense

    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
    heim
    -0.11
    ä¸Ģ代
    -0.09
    ertools
    -0.09
    çĽij
    -0.09
    åŃIJéĩĮ
    -0.09
    æĢĻ
    -0.08
    å±ŀäºİèĩªå·±
    -0.08
    çĶļèĩ³è¿ĺ
    -0.08
    ä¸ĢåıĮ
    -0.08
    ading
    -0.08
    POSITIVE LOGITS
    бÑĢо
    0.09
     laten
    0.09
    æ¦Ń
    0.08
    -leaning
    0.08
    Ñħод
    0.08
    åijĺ
    0.08
     Rust
    0.08
    åı£ç¢ij
    0.07
    诩
    0.07
    óc
    0.07
    Activations Density 0.004%

    No Known Activations