© 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. 11342
    Prev
    Next
    INDEX
    Explanations

    He followed by names or titles

    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
    science
    -0.10
    ski
    -0.09
    ãģĦãģ¾ãģĻ
    -0.09
    lier
    -0.09
    NewLabel
    -0.08
    illo
    -0.08
    ç¢į
    -0.08
    ä»Ĭ天æĪij们
    -0.08
    nya
    -0.08
    é¦ĸåħĪè¦ģ
    -0.08
    POSITIVE LOGITS
    oretical
    0.12
    er
    0.11
    aney
    0.11
    uristic
    0.11
    idelberg
    0.10
    asley
    0.10
    缮åħī
    0.10
    aley
    0.10
    inerary
    0.10
    est
    0.09
    Activations Density 0.051%

    No Known Activations