© 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. 39113
    Prev
    Next
    INDEX
    Explanations

    data formats or representation

    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
    è´¯
    -0.09
    Unload
    -0.09
    è½½
    -0.08
    ãĥ«ãĥī
    -0.08
    Sad
    -0.08
    KEN
    -0.08
    daf
    -0.08
     pylint
    -0.08
     rod
    -0.08
    æ±
    -0.08
    POSITIVE LOGITS
    æĽ´æľī
    0.10
     separate
    0.10
    éĤ£æł·
    0.09
    æĪIJåĵģ
    0.09
    ä¸Ģ个个
    0.09
    éĩįä¸Ńä¹ĭéĩį
    0.09
    ä»¿ä½Ľ
    0.09
    çī¹æ®ĬæĥħåĨµ
    0.09
    cept
    0.09
    çĥ«
    0.09
    Activations Density 0.176%

    No Known Activations