© 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. 4501
    Prev
    Next
    INDEX
    Explanations

    Copyright notices or identifiers

    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
    ings
    -0.14
    ìĿ¸íĦ°
    -0.10
    akah
    -0.10
    tery
    -0.10
    ils
    -0.09
    ìŀ¡
    -0.09
    ä¸įå¿į
    -0.09
    ween
    -0.09
    饴
    -0.09
    å¥Ĥ
    -0.09
    POSITIVE LOGITS
     /*<<<
    0.13
    perial
    0.13
    olated
    0.13
    olate
    0.12
    ividual
    0.12
    mediately
    0.11
    itial
    0.11
     iVar
    0.10
    æļĩ
    0.10
    /out
    0.10
    Activations Density 1.939%

    No Known Activations