© 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. 55805
    Prev
    Next
    INDEX
    Explanations

    expectation or expected

    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
    åĪ»
    -0.09
    gue
    -0.09
    focused
    -0.08
    _qos
    -0.08
    欢
    -0.08
    è¿Ļåľº
    -0.08
    å°¤
    -0.08
    èĩªè±ª
    -0.08
     \/
    -0.08
    Gu
    -0.08
    POSITIVE LOGITS
    agli
    0.10
    åģļçļĦäºĭæĥħ
    0.10
    积æŀģåıĤä¸İ
    0.10
    åłĤ
    0.10
    lesh
    0.09
    伸æīĭ
    0.09
    _('
    0.09
    ovic
    0.09
    ulk
    0.09
    æīĭèĦļ
    0.09
    Activations Density 0.020%

    No Known Activations