© 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. 61145
    Prev
    Next
    INDEX
    Explanations

    reflex actions

    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
    ä¸īåĪĨä¹ĭä¸Ģ
    -0.09
    çĨŁ
    -0.09
    åıįæŃ£
    -0.09
    forced
    -0.09
    uments
    -0.09
    ĮĢ
    -0.09
    uration
    -0.09
    ç²¹
    -0.09
    èĻ±
    -0.08
    ady
    -0.08
    POSITIVE LOGITS
    ively
    0.11
    åŁİå¸Ĥåıijå±ķ
    0.11
    ../../../
    0.09
    es
    0.09
    éªĭ
    0.09
    _dash
    0.09
    饮
    0.09
    eds
    0.09
    itas
    0.09
    дов
    0.09
    Activations Density 0.006%

    No Known Activations