© 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. 57730
    Prev
    Next
    INDEX
    Explanations

    dip and dips

    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
    çĮ·
    -0.10
    åħīçݯ
    -0.09
    สรร
    -0.09
    Ø´ÙĤ
    -0.09
    utsch
    -0.09
    èĥ¼
    -0.09
    /tutorial
    -0.09
    è½°
    -0.08
    å´©
    -0.08
    æĢİä¹Īèĥ½
    -0.08
    POSITIVE LOGITS
     dip
    0.11
     dipped
    0.10
     Dip
    0.10
    åį®
    0.10
     dips
    0.09
     cue
    0.09
     dipping
    0.09
     into
    0.09
    篮
    0.08
    (/^
    0.08
    Activations Density 0.006%

    No Known Activations