© 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. 50291
    Prev
    Next
    INDEX
    Explanations

    inquiry

    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
    omer
    -0.10
    åĦ¿
    -0.10
    ä¸Ģå¥Ĺ
    -0.09
    à¹Ĩ
    -0.09
     Ñģобой
    -0.08
    ember
    -0.08
    æ°Ĺãģ¥
    -0.08
    ening
    -0.08
    ãģĩ
    -0.08
    çĺ¾
    -0.08
    POSITIVE LOGITS
    ìĤ¬íķŃ
    0.14
    ingly
    0.12
    颦
    0.09
    صر
    0.09
    edis
    0.09
    annis
    0.09
    uset
    0.09
    ubits
    0.09
    веÑĢ
    0.09
    вали
    0.09
    Activations Density 0.027%

    No Known Activations