© 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. 57845
    Prev
    Next
    INDEX
    Explanations

    average

    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
    ÑĢак
    -0.10
    æĥ´
    -0.09
    æĹ©æĹ©
    -0.09
    å®Įåħ¨åı¯ä»¥
    -0.09
    -ie
    -0.09
    astes
    -0.09
    éĺĨ
    -0.09
    èĢ¨
    -0.09
    æ´¾éģ£
    -0.09
    JUST
    -0.09
    POSITIVE LOGITS
    å¹³åĿĩ
    0.20
     average
    0.19
    average
    0.17
    å¹³åĿĩæ°´å¹³
    0.16
    Average
    0.15
    _average
    0.15
     Average
    0.15
     avg
    0.14
    .average
    0.13
     aver
    0.13
    Activations Density 0.098%

    No Known Activations