© 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. 3675
    Prev
    Next
    INDEX
    Explanations

    can't action verb

    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
    æĥ¯ä¾ĭ
    -0.10
    æķ¢
    -0.10
    èĥĥåı£
    -0.09
    讵
    -0.09
    èĥ½å¤Łè®©
    -0.09
    ited
    -0.09
    æĶ»åĬ¿
    -0.09
     наиболее
    -0.08
    hawk
    -0.08
    çīĪæĿĥæīĢæľī
    -0.08
    POSITIVE LOGITS
    onical
    0.13
     бÑĭÑĤÑĮ
    0.10
    /w
    0.10
    èĥľä»»
    0.10
    berra
    0.10
    åģļåĪ°
    0.09
    tered
    0.09
    çe
    0.09
    -être
    0.09
    ç®Ĺæĺ¯
    0.09
    Activations Density 0.026%

    No Known Activations