© 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. 17365
    Prev
    Next
    INDEX
    Explanations

    rule of

    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
    å¤įå·¥
    -0.10
    æľīåĪ©
    -0.09
    dress
    -0.09
    è¿Ļ份
    -0.09
    ovid
    -0.09
    aben
    -0.09
     hạng
    -0.09
    衬
    -0.09
    _IO
    -0.08
    çīĩ
    -0.08
    POSITIVE LOGITS
     Rule
    0.11
    (rule
    0.11
     rule
    0.10
    rule
    0.10
    Rule
    0.10
    RULE
    0.09
    -rule
    0.09
    å°º
    0.09
     Dominic
    0.08
     Nur
    0.08
    Activations Density 0.061%

    No Known Activations