© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38155
    Prev
    Next
    INDEX
    Explanations

    shortly after

    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
     Sachs
    -0.10
    ç²¹
    -0.09
    çģ«çĤ¬
    -0.09
    thal
    -0.09
    Ñĥ
    -0.08
    àµį
    -0.08
    EMU
    -0.08
    çıij
    -0.08
    çģ«çĥ§
    -0.08
     ster
    -0.08
    POSITIVE LOGITS
    zeitig
    0.12
    便åı¯
    0.10
    chy
    0.10
    ä¸įä¹ħ
    0.10
    çĶļèĩ³æĺ¯
    0.10
    comings
    0.10
    itect
    0.09
    ãģŀ
    0.09
    immer
    0.09
    itude
    0.09
    Activations Density 0.005%

    No Known Activations