© 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. 55334
    Prev
    Next
    INDEX
    Explanations

    economical, economy, efficiency, mileage

    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
    _rwlock
    -0.09
    æĭŁ
    -0.09
    èĥ³
    -0.09
    æĪĺäºī
    -0.09
    irement
    -0.09
    visa
    -0.08
     Ñįк
    -0.08
    akedown
    -0.08
    Ĥæķ°
    -0.08
    åĮºåĨħ
    -0.08
    POSITIVE LOGITS
    trag
    0.10
    羸
    0.09
     onload
    0.09
    .Chain
    0.08
    è§Ħ
    0.08
    ç©·
    0.08
    èĻ¾
    0.08
     chá
    0.08
    å¿ĥæĦ¿
    0.08
     بÙĨÙģس
    0.08
    Activations Density 0.009%

    No Known Activations