© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 60898
    Prev
    Next
    INDEX
    Explanations

    obvious configurations

    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
    çĮĹ
    -0.10
    çħ
    -0.09
    fully
    -0.09
    ÑĢай
    -0.08
    erin
    -0.08
    angs
    -0.08
    pent
    -0.08
    çĿĢä»ĸ
    -0.08
    æłĩåĩĨ
    -0.08
    ãģ£ãģįãĤĬ
    -0.08
    POSITIVE LOGITS
    ness
    0.16
    ity
    0.12
    -eyed
    0.10
    ities
    0.10
    ance
    0.10
    ment
    0.10
    -looking
    0.10
    è¿Ļä¸ĢçĤ¹
    0.09
    èĢĮåıĪ
    0.09
    ITY
    0.09
    Activations Density 0.013%

    No Known Activations