© 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. 1216
    Prev
    Next
    INDEX
    Explanations

    ladder, stairs, staircase

    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
    æĪ½
    -0.10
    ishi
    -0.09
    elves
    -0.08
     дела
    -0.08
    åĶ§
    -0.08
    anni
    -0.08
    nummer
    -0.08
    å¤§åľ°
    -0.08
    pra
    -0.08
    ure
    -0.08
    POSITIVE LOGITS
    ê¼´
    0.10
    èµ°ä¸ĭåİ»
    0.10
    绿åŁİ
    0.09
    маÑĤеÑĢиал
    0.09
    combe
    0.09
    IFORM
    0.09
    iras
    0.09
     ladder
    0.09
    /bus
    0.08
    ritt
    0.08
    Activations Density 0.011%

    No Known Activations