© 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. 19960
    Prev
    Next
    INDEX
    Explanations

    units of measurement

    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
    íݸ
    -0.09
    å¯¹æľªæĿ¥
    -0.09
    stk
    -0.09
    æİIJ
    -0.09
    fort
    -0.09
    alach
    -0.09
    èĥ¸
    -0.08
    ç½¢äºĨ
    -0.08
    iri
    -0.08
    ç®ĹäºĨ
    -0.08
    POSITIVE LOGITS
    åĪĨæĶ¯
    0.09
    _HAL
    0.09
    Blob
    0.08
    èij«
    0.08
    uzzi
    0.08
    è¦ģçĶ¨
    0.08
    Hal
    0.08
    /tools
    0.08
    اÙģت
    0.08
    ë°į
    0.08
    Activations Density 0.069%

    No Known Activations