© 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. 23378
    Prev
    Next
    INDEX
    Explanations

    body parts and prepositions

    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
    _OPENGL
    -0.09
    .val
    -0.09
    ä»İä¸Ń
    -0.09
    å¯Į
    -0.09
    塬
    -0.09
    angelo
    -0.09
    èģĶè°Ĭ
    -0.09
    lessly
    -0.09
     Staples
    -0.08
    åľ¨æѤä¹ĭåīį
    -0.08
    POSITIVE LOGITS
    åĮħ袱
    0.10
    æµĬ
    0.09
    çĦī
    0.08
    èµĥ
    0.08
    vd
    0.08
    TIM
    0.08
    好åIJĹ
    0.08
    äºĨå¾Īä¹ħ
    0.08
    èĪĴæľį
    0.08
    åħ¥
    0.08
    Activations Density 0.009%

    No Known Activations