© 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. 54320
    Prev
    Next
    INDEX
    Explanations

    conditioning or conditioned

    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
    /close
    -0.10
    istic
    -0.09
    ALLED
    -0.08
    èĮĮ
    -0.08
    aina
    -0.08
    ugh
    -0.08
    æ¢Ń
    -0.08
    è½½
    -0.08
    ocha
    -0.08
    /umd
    -0.08
    POSITIVE LOGITS
    åĽ½äºº
    0.10
    erde
    0.09
    als
    0.09
    è¿ĩåī©
    0.09
    ç®į
    0.09
    ERA
    0.09
    åįģ足
    0.09
    ãģĦãģ¤
    0.09
    arf
    0.09
    uyen
    0.08
    Activations Density 0.037%

    No Known Activations