© 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. 17434
    Prev
    Next
    INDEX
    Explanations

    were/are/have + past participle

    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
    è§
    -0.09
     ae
    -0.09
    饱
    -0.08
    æįº
    -0.08
    éĩľ
    -0.08
    ò
    -0.08
    ТЬ
    -0.08
    ä¸Ģåī¯
    -0.08
    绸
    -0.08
    屯
    -0.08
    POSITIVE LOGITS
    acement
    0.08
    udi
    0.08
    è°
    0.08
    rais
    0.08
    åŃľ
    0.08
    alus
    0.08
     Giul
    0.08
    raise
    0.08
    好好
    0.07
    _pas
    0.07
    Activations Density 0.088%

    No Known Activations