© 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. 32449
    Prev
    Next
    INDEX
    Explanations

    canoe, rowing, K-2

    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
    æ¯Ĵ
    -0.10
    宦
    -0.10
    lust
    -0.09
    峪
    -0.09
    åĬ©
    -0.09
    å±¥
    -0.09
    ä»ķ
    -0.09
    azard
    -0.09
    arez
    -0.09
    esimal
    -0.09
    POSITIVE LOGITS
    æ¼Ĥæµģ
    0.14
     canoe
    0.10
    桨
    0.09
    èĪ¹åıª
    0.09
    渡
    0.09
    èį»
    0.09
    æ¸Ķ
    0.09
    禧
    0.09
    lever
    0.09
    æīĢæľī人
    0.09
    Activations Density 0.123%

    No Known Activations