© 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. 30671
    Prev
    Next
    INDEX
    Explanations

    generated, distributed, downloaded, originally

    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
    ively
    -0.12
     currently
    -0.11
    缮åīį
    -0.11
    ened
    -0.11
    ãģłãĤįãģĨ
    -0.10
    缮åīįæŃ£åľ¨
    -0.10
    currently
    -0.10
    缮åīį为æŃ¢
    -0.09
    elman
    -0.09
    çļĤ
    -0.09
    POSITIVE LOGITS
    nt
    0.20
    /is
    0.20
     originally
    0.12
    origin
    0.12
    ãģģ
    0.11
    ness
    0.11
    ÑĭÑĪ
    0.11
     ÑĢанее
    0.11
     бÑĭ
    0.10
    /am
    0.10
    Activations Density 0.458%

    No Known Activations