© 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. 63450
    Prev
    Next
    INDEX
    Explanations

    introduction signals

    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
     ‘
    -0.14
    ‘
    -0.11
    =”
    -0.10
     ’
    -0.10
    ’s
    -0.10
     “
    -0.09
    ’re
    -0.09
    ’d
    -0.09
    ’t
    -0.09
    ’ve
    -0.09
    POSITIVE LOGITS
    ing
    0.13
    Ø©
    0.11
    า
    0.10
    åľ¨è¿Ļç§į
    0.09
     Ramos
    0.09
    errupt
    0.09
    æ±ģ
    0.09
    ryption
    0.09
    ropri
    0.09
    ä»Ģä¹ĪåİŁåĽł
    0.09
    Activations Density 0.136%

    No Known Activations