© 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. 33197
    Prev
    Next
    INDEX
    Explanations

    move up or out of bed

    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
    åģ¶
    -0.09
    紧紧
    -0.08
    她们
    -0.08
    å½»åºķ
    -0.08
    èĭ¹
    -0.08
     Blake
    -0.08
     Ping
    -0.08
     Ñģво
    -0.08
     Laws
    -0.08
    æ³ķè§Ħ
    -0.08
    POSITIVE LOGITS
    è·³åĩº
    0.11
    éĨĴæĿ¥
    0.10
    起身
    0.10
    èµ·åºĬ
    0.10
    stants
    0.10
    onders
    0.09
    ãģĭãģ¤
    0.09
    eon
    0.09
    åģ·
    0.09
    æĹĭ
    0.08
    Activations Density 0.020%

    No Known Activations