© 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. 33980
    Prev
    Next
    INDEX
    Explanations

    uniform followed by resource, date, rate, or magnetic

    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
    falls
    -0.09
    LONG
    -0.09
    -0.09
    åıªèĥ½è¯´
    -0.09
    олÑĮко
    -0.09
    çĹĴ
    -0.09
    IFF
    -0.09
    issors
    -0.09
    ç²¥
    -0.09
    ç²ī
    -0.09
    POSITIVE LOGITS
    ity
    0.17
    sWith
    0.13
    emente
    0.12
    ally
    0.11
    itarian
    0.11
    ãģĹãģ¾ãģĨ
    0.11
    cy
    0.11
     xuyên
    0.10
    ality
    0.10
    ers
    0.09
    Activations Density 0.020%

    No Known Activations