© 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. 3094
    Prev
    Next
    INDEX
    Explanations

    home directories and related contexts

    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
    -0.14
    unami
    -0.11
    à¹Į
    -0.11
    ments
    -0.10
    edImage
    -0.10
    ãģĦãģŁãĤī
    -0.09
    -0.09
    ive
    -0.09
    ured
    -0.09
    marsh
    -0.09
    POSITIVE LOGITS
    grown
    0.15
    brew
    0.14
    coming
    0.13
    stead
    0.13
    lessness
    0.12
    bei
    0.11
    less
    0.11
    iliar
    0.11
    lessly
    0.10
    ำ
    0.09
    Activations Density 0.062%

    No Known Activations