© 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. 2844
    Prev
    Next
    INDEX
    Explanations

    cell, free, one, software, portfolio, guide, future, data, stress, complexity

    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
    -0.11
    沧
    -0.11
    以éĺ²
    -0.11
    绸
    -0.10
    éĢį
    -0.10
    æĪij们ä¸Ģèµ·
    -0.09
    è·ij
    -0.09
    èĽŁ
    -0.09
    毯
    -0.09
    çħ§äº®
    -0.09
    POSITIVE LOGITS
    ä¹ĥ
    0.10
    rastructure
    0.10
     curt
    0.08
    istles
    0.08
    俪
    0.07
     way
    0.07
    CUR
    0.07
    eller
    0.07
    chs
    0.07
    anches
    0.07
    Activations Density 1.592%

    No Known Activations