© 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. 33860
    Prev
    Next
    INDEX
    Explanations

    implementation

    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
     Legacy
    -0.09
    pak
    -0.09
     Civic
    -0.09
    arest
    -0.08
    èѬ
    -0.08
    ched
    -0.08
    æłī
    -0.08
    çĵ¢
    -0.08
    å°±æѤ
    -0.08
    ä¸Ģè¾ĪåŃIJ
    -0.08
    POSITIVE LOGITS
     ÑĢеализаÑĨи
    0.12
    enza
    0.10
    å¥ī
    0.10
    plementation
    0.09
     implementations
    0.09
     виде
    0.09
    ราย
    0.09
    ensa
    0.09
    за
    0.09
     implementation
    0.09
    Activations Density 0.165%

    No Known Activations