© 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. 31435
    Prev
    Next
    INDEX
    Explanations

    at whatever cost

    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
    åĽłä¸ºæĪij
    -0.10
    æĪijåıªæĺ¯
    -0.09
    æĮ£éĴ±
    -0.09
    çĪ±åIJĥ
    -0.09
    åĸľæ¬¢åIJĥ
    -0.09
    åĽłçĤº
    -0.09
    渥
    -0.09
    çŃīçĿĢ
    -0.09
    rame
    -0.09
     bankers
    -0.08
    POSITIVE LOGITS
    çīºçī²
    0.15
    ä¸įæĥľ
    0.11
    åĨĴéĻ©
    0.10
    çľīæ¯Ľ
    0.10
    忽çķ¥äºĨ
    0.10
    çīº
    0.10
    outu
    0.10
    å¾Ģå¾Ģä¼ļ
    0.09
    为æѤ
    0.09
    çľī
    0.09
    Activations Density 0.146%

    No Known Activations