© 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. 33308
    Prev
    Next
    INDEX
    Explanations

    remote locations

    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
    黳
    -0.09
    çĽĪ
    -0.09
     plaza
    -0.09
    å°ıåIJĥ
    -0.09
    åĽ´ç»ķ
    -0.09
    ngine
    -0.09
    çĭ±
    -0.09
    åħļç»Ħ
    -0.09
     mantle
    -0.09
    è°Ħ
    -0.09
    POSITIVE LOGITS
    åĥ»
    0.19
    isol
    0.18
    åģıè¿ľ
    0.18
    åѤç«ĭ
    0.17
     isolated
    0.17
    éļĶ离
    0.17
     isolation
    0.16
    è¿ľç¦»
    0.16
    åѤ
    0.15
    åѤçĭ¬
    0.14
    Activations Density 0.087%

    No Known Activations