© 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. 49700
    Prev
    Next
    INDEX
    Explanations

    optical communication

    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
    abouts
    -0.12
    asing
    -0.09
    ement
    -0.09
    inee
    -0.09
    istration
    -0.09
    èĢħçļĦ
    -0.09
    好ä¸į容æĺĵ
    -0.09
    hard
    -0.09
    çĮ¾
    -0.09
    çĹĽçĤ¹
    -0.08
    POSITIVE LOGITS
    imal
    0.11
    çĦ¶æĺ¯
    0.09
    ÑĢÑĥд
    0.09
    箸
    0.08
    ä»İ严
    0.08
    eam
    0.08
    æĪIJ人
    0.08
    ights
    0.08
    ting
    0.08
    ight
    0.08
    Activations Density 0.066%

    No Known Activations