© 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. 11408
    Prev
    Next
    INDEX
    Explanations

    names ending in lith/itch/dale/ville

    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
    è¹Ī
    -0.09
    kes
    -0.09
    豫
    -0.09
    inch
    -0.09
    åľ°æĸ¹
    -0.08
    AVED
    -0.08
    çĦĻ
    -0.08
    ãĥĥãĤ¯ãĤ¹
    -0.08
    mdl
    -0.08
    mods
    -0.08
    POSITIVE LOGITS
     lith
    0.10
    å··
    0.10
     Lith
    0.09
    çī¢
    0.08
    æį£
    0.08
     lit
    0.08
    éģ´
    0.08
    åĺĪ
    0.08
    lit
    0.08
    æĸ¯
    0.08
    Activations Density 0.063%

    No Known Activations