© 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. 18570
    Prev
    Next
    INDEX
    Explanations

    turbulence, turmoil, instability, volatile

    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
    çľī
    -0.10
    彬
    -0.09
    ias
    -0.09
    讳
    -0.08
    avor
    -0.08
    鹿
    -0.08
    ting
    -0.08
    IAS
    -0.08
    killer
    -0.08
    壤
    -0.08
    POSITIVE LOGITS
    ä¸įå®ī
    0.18
    åĬ¨èį¡
    0.15
    ä¸įå®ļ
    0.13
    ä¸įåłª
    0.12
    ä¸į稳å®ļ
    0.12
    å±ĢåĬ¿
    0.10
    ä¸įç¡®å®ļ
    0.10
    åħ¶éĹ´
    0.10
    åµĺ
    0.09
    ä¸įæ¸ħ
    0.09
    Activations Density 0.031%

    No Known Activations