© 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. 39914
    Prev
    Next
    INDEX
    Explanations

    forehead and frontal

    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
    åıijå±ķæ½ľåĬĽ
    -0.10
    athon
    -0.09
    ERO
    -0.09
    neck
    -0.09
    责任å¿ĥ
    -0.09
    æĿ¥åĪ°äºĨ
    -0.09
    æ¼ī
    -0.09
    æĻļæĬ¥
    -0.09
    çģ«çĥ§
    -0.08
    æ¾
    -0.08
    POSITIVE LOGITS
     bande
    0.09
    å±ħ室
    0.09
    åħ«åŃĹ
    0.09
    aly
    0.09
    borg
    0.08
     gyr
    0.08
    æĺı
    0.08
    achable
    0.08
    ç¥ĸ
    0.08
    once
    0.08
    Activations Density 0.024%

    No Known Activations