© 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. 6497
    Prev
    Next
    INDEX
    Explanations

    not only, but

    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
     zwar
    -0.09
    GLOBALS
    -0.08
    hook
    -0.08
    ãģ§ãģįãģ¦
    -0.08
    ä¹Ī
    -0.08
    merc
    -0.08
    èĩªçĦ¶èĢĮ
    -0.08
    ting
    -0.08
    人们对
    -0.08
    atel
    -0.07
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.14
     sondern
    0.11
    loom
    0.10
    å°±è¿ŀ
    0.10
    utzer
    0.09
    uits
    0.09
    aped
    0.09
    еÑĩа
    0.08
     dess
    0.08
    ãģ£ãģŁãĤī
    0.08
    Activations Density 0.159%

    No Known Activations