© 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. 56720
    Prev
    Next
    INDEX
    Explanations

    mood and emotional states

    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
    æ§
    -0.09
    èĪĮ
    -0.09
    seg
    -0.08
     ih
    -0.08
     Guard
    -0.08
    èĪĶ
    -0.08
    Assert
    -0.08
    ès
    -0.08
    egrity
    -0.08
     dit
    -0.07
    POSITIVE LOGITS
    à¸ĵ
    0.12
    Fuse
    0.12
    俦
    0.12
    emachine
    0.10
    åĩºéŨ
    0.10
    -metadata
    0.10
     jä
    0.09
     кли
    0.09
    管çIJĨæĿ¡ä¾ĭ
    0.09
    æģŃåĸľ
    0.09
    Activations Density 0.025%

    No Known Activations