© 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. 33169
    Prev
    Next
    INDEX
    Explanations

    variation and variability

    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¸IJ
    -0.09
    å¾Ĺ天
    -0.08
    é¦Ī
    -0.08
    ubar
    -0.08
    çī¢åĽº
    -0.08
    æ¯ħ
    -0.07
    èĬĴ
    -0.07
    趣
    -0.07
    eos
    -0.07
    æİł
    -0.07
    POSITIVE LOGITS
     variation
    0.15
     variability
    0.15
    åıĺå¼Ĥ
    0.14
     Variation
    0.13
    variation
    0.13
    æ³¢åĬ¨
    0.12
    çݰ象
    0.12
    _variation
    0.11
     ваÑĢи
    0.11
    andard
    0.11
    Activations Density 0.311%

    No Known Activations