© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52618
    Prev
    Next
    INDEX
    Explanations

    therapeutic targets

    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
    _MOUNT
    -0.09
    æŃ£éĿ¢
    -0.09
    åĪ©çĽĬ
    -0.08
    Ť
    -0.08
    igar
    -0.08
    eer
    -0.08
    è¾ħ导
    -0.08
    QA
    -0.08
    çļĦäºĭ
    -0.08
    游åĩ»
    -0.08
    POSITIVE LOGITS
    æĭīä¸ģ
    0.10
    åģ¶
    0.09
    ç®Ģ约
    0.09
    大å®ĹåķĨåĵģ
    0.09
     Cumberland
    0.08
    ÑĦа
    0.08
    ORY
    0.08
    åĨĻåĩº
    0.08
    KeyEvent
    0.08
    .listeners
    0.08
    Activations Density 0.142%

    No Known Activations