© 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. 18386
    Prev
    Next
    INDEX
    Explanations

    distributed with this work

    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
    åŃIJéĩĮ
    -0.12
    ราย
    -0.10
    icom
    -0.10
     workforce
    -0.10
    ities
    -0.10
    nist
    -0.10
    à¸ĩาà¸Ļ
    -0.09
    vil
    -0.09
    å·¥ä½ľæľºåĪ¶
    -0.09
    å·¥ä½ľäººåijĺ
    -0.09
    POSITIVE LOGITS
    fully
    0.16
    shops
    0.15
    loads
    0.13
    stations
    0.13
    horse
    0.13
    ings
    0.12
    lessness
    0.12
    lessly
    0.11
    around
    0.11
    bench
    0.11
    Activations Density 0.165%

    No Known Activations