© 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. 54102
    Prev
    Next
    INDEX
    Explanations

    water canals, irrigation, drainage

    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
    odyn
    -0.11
    .getActivity
    -0.10
    大åݦ
    -0.09
    åݦ
    -0.09
    山路
    -0.09
    _until
    -0.09
    ä¸Ģæŀļ
    -0.08
    woods
    -0.08
    çĢijå¸ĥ
    -0.08
    伺æľį
    -0.08
    POSITIVE LOGITS
    ickness
    0.10
    躺
    0.09
    shal
    0.09
    强åĪ¶
    0.09
    ISH
    0.08
    è¹Ļ
    0.08
     Autof
    0.08
     Chick
    0.08
    èĬĴ
    0.08
     Dud
    0.08
    Activations Density 0.073%

    No Known Activations