© 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. 56098
    Prev
    Next
    INDEX
    Explanations

    spur followed by development, investigations, or moment

    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
    ario
    -0.09
    ari
    -0.09
    agner
    -0.09
    akes
    -0.08
    ting
    -0.08
    uration
    -0.08
    åħ¬å¼Ģåıijè¡Į
    -0.08
    æĢģ度
    -0.08
    оÑĢо
    -0.08
    顺
    -0.08
    POSITIVE LOGITS
    lessly
    0.13
    ingly
    0.11
    iously
    0.10
    å·§åIJĪ
    0.10
    nett
    0.09
    CRT
    0.09
    ../../../
    0.09
    fully
    0.09
    ä¸įå·²
    0.09
    hart
    0.09
    Activations Density 0.004%

    No Known Activations