© 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. 63590
    Prev
    Next
    INDEX
    Explanations

    intake of

    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
    UnderTest
    -0.09
    潺
    -0.08
    ieur
    -0.08
    lijk
    -0.08
     Ring
    -0.08
    miss
    -0.08
    çĦ¶æĺ¯
    -0.08
    ally
    -0.08
    ly
    -0.08
    mente
    -0.08
    POSITIVE LOGITS
    eing
    0.11
    ez
    0.10
    /release
    0.10
    ean
    0.10
    èİ
    0.10
    visor
    0.10
    onium
    0.10
    æĸ¯é¡¿
    0.09
    ornado
    0.09
    clusive
    0.09
    Activations Density 0.011%

    No Known Activations