© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36457
    Prev
    Next
    INDEX
    Explanations

    copyright notices or code formatting

    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
    	TokenName
    -0.10
    大æ´ĭ
    -0.09
    åħļå·¥å§Ķ
    -0.09
    åľŁèĢ³åħ¶
    -0.09
    iena
    -0.09
     dużo
    -0.09
    ptions
    -0.08
    æIJª
    -0.08
    tier
    -0.08
    moire
    -0.08
    POSITIVE LOGITS
    onomy
    0.12
    perature
    0.10
    phoon
    0.10
    orarily
    0.10
    pson
    0.10
    uesday
    0.10
    .currentThread
    0.09
    imonial
    0.09
    quila
    0.09
    жд
    0.09
    Activations Density 1.122%

    No Known Activations