© 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. 23140
    Prev
    Next
    INDEX
    Explanations

    I think

    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
    ebra
    -0.09
     guess
    -0.09
    ulg
    -0.08
    otype
    -0.08
     Guess
    -0.08
    éļı便
    -0.08
    guess
    -0.08
    两边
    -0.08
    distributed
    -0.08
    4
    -0.07
    POSITIVE LOGITS
    ä¸Ĭå¹´
    0.12
    说è¿ĩ
    0.10
    æĽ¾åľ¨
    0.10
    æ±
    0.10
    è·Ľ
    0.10
    è§ģè¿ĩ
    0.09
    æįĭ
    0.09
    å¤ĸåľ°
    0.09
    AVED
    0.09
    æĮ½æķij
    0.09
    Activations Density 0.100%

    No Known Activations