© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7387
    Prev
    Next
    INDEX
    Explanations

    exclamation mark

    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
    hte
    -0.09
     diffuse
    -0.09
    ays
    -0.09
    казан
    -0.09
    _zeros
    -0.08
    çķª
    -0.08
    ëĦIJ
    -0.08
    ãģ¹
    -0.08
    ÑģÑĤавка
    -0.08
    åħ³éĶ®åŃĹ
    -0.08
    POSITIVE LOGITS
    eczy
    0.09
    ä¸į解
    0.09
    plode
    0.09
    éķŃ
    0.09
    åĽ½æľīèµĦ产
    0.09
    å¼Ħ
    0.09
    ForResource
    0.09
    ä¿Ŀå¯Ĩ
    0.09
    åĢº
    0.08
    ç¥ŀç§ĺ
    0.08
    Activations Density 0.001%

    No Known Activations