© 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. 6399
    Prev
    Next
    INDEX
    Explanations

    amounts of data

    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
    å¤ļäºĨ
    -0.09
    常çĶ¨
    -0.09
    äºĨä¸Ģåľº
    -0.09
    éĩį大
    -0.09
    éĻIJ度
    -0.09
    ä»»ä½ķæŶåĢĻ
    -0.08
    äºĨä¸Ģç³»åĪĹ
    -0.08
    ATUS
    -0.08
    çĻ¾èĬ±
    -0.08
    ered
    -0.08
    POSITIVE LOGITS
     paperwork
    0.10
    æį
    0.10
    ä¼ĺç§ĢçļĦ
    0.09
    ç²¾åĬĽ
    0.09
    dana
    0.09
    thest
    0.08
    ä¸įçĦ¶
    0.08
    èģĮä¸ļæĬĢæľ¯
    0.08
    ataka
    0.08
    uded
    0.08
    Activations Density 0.050%

    No Known Activations