© 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. 35404
    Prev
    Next
    INDEX
    Explanations

    fire starting or initiating

    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
    лÑıÑĢ
    -0.10
    æ´Ĺæīĭ
    -0.09
    åĬłçĽŁåķĨ
    -0.09
    ç´§è·Ł
    -0.09
     Blades
    -0.09
    REEN
    -0.09
    Ñıн
    -0.09
    sey
    -0.08
    æĻ¤
    -0.08
    大å±Ģ
    -0.08
    POSITIVE LOGITS
     Fired
    0.09
    xed
    0.09
    .Fire
    0.09
    nze
    0.09
    å¼ĢæĿ¥
    0.09
     rockets
    0.09
     fired
    0.09
    黾
    0.08
    /pass
    0.08
    ned
    0.08
    Activations Density 0.017%

    No Known Activations