© 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. 34158
    Prev
    Next
    INDEX
    Explanations

    sodium chloride

    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
    çģ«çĤ¬
    -0.11
     Gaz
    -0.09
    овоÑĢ
    -0.08
    æĨĤ
    -0.08
    à¸ķร
    -0.08
     Torch
    -0.08
    çĹ¿
    -0.08
    дей
    -0.08
    mdb
    -0.08
    èIJ¥éĢłèī¯å¥½
    -0.08
    POSITIVE LOGITS
    转载请
    0.12
    éĴł
    0.11
    Trait
    0.10
     sodium
    0.10
     salt
    0.09
    çĪ¸
    0.09
    è´¢æĶ¿
    0.09
    çĪ¶åŃIJ
    0.09
    rollo
    0.09
    Dash
    0.08
    Activations Density 0.042%

    No Known Activations