© 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. 17402
    Prev
    Next
    INDEX
    Explanations

    skip when copying or between groups

    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
    ÙĦÙģ
    -0.11
    uite
    -0.10
    ãģıãĤĭ
    -0.09
    ullo
    -0.09
    sWith
    -0.09
    croft
    -0.09
    éĤij
    -0.09
    견
    -0.09
    rowse
    -0.08
    angement
    -0.08
    POSITIVE LOGITS
    pered
    0.12
    per
    0.11
    pery
    0.11
    listed
    0.10
    ellaneous
    0.10
    aroo
    0.10
    ä½Ĩæĺ¯å¯¹äºİ
    0.09
    æīĪ
    0.09
    çŁ¾
    0.09
    jack
    0.09
    Activations Density 0.016%

    No Known Activations