© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37696
    Prev
    Next
    INDEX
    Explanations

    child abuse and maltreatment

    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
    _SCL
    -0.09
    .setAction
    -0.08
    reib
    -0.08
    .setType
    -0.08
    ç¥ľ
    -0.08
    ffi
    -0.08
    à¦ĩ
    -0.08
    éĴ±å¸ģ
    -0.08
    _prepare
    -0.08
    çĸ¡
    -0.08
    POSITIVE LOGITS
    aÄĩ
    0.10
    hood
    0.09
    iqu
    0.09
    aise
    0.09
    aire
    0.09
     Vox
    0.08
    ÙĬÙĩ
    0.08
    -l
    0.08
    care
    0.08
    ishly
    0.08
    Activations Density 0.027%

    No Known Activations