© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9610
    Prev
    Next
    INDEX
    Explanations

    similar concepts

    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
    amples
    -0.10
    nable
    -0.10
    olis
    -0.10
    ä¸Ģ代
    -0.10
    nell
    -0.09
    agt
    -0.09
    ammer
    -0.09
    ÑĭÑĪ
    -0.09
    uristic
    -0.09
    çłĴ
    -0.09
    POSITIVE LOGITS
    -minded
    0.15
    ily
    0.15
    -sized
    0.15
    ities
    0.14
    mente
    0.14
     minded
    0.13
    -aged
    0.12
     sized
    0.12
    emente
    0.11
    -sex
    0.11
    Activations Density 0.026%

    No Known Activations