© 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. 23743
    Prev
    Next
    INDEX
    Explanations

    instance of

    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
    -0.13
    inton
    -0.11
    thing
    -0.10
    oment
    -0.10
    INGTON
    -0.09
    çĪ¶åŃIJ
    -0.09
    thon
    -0.09
    ston
    -0.09
    strap
    -0.09
    paring
    -0.09
    POSITIVE LOGITS
    aneously
    0.17
    aneous
    0.13
    ytut
    0.12
    ively
    0.11
    ally
    0.11
    ê¹IJ
    0.10
    æĥķ
    0.10
    utors
    0.10
    пÑĥ
    0.10
    wit
    0.10
    Activations Density 0.042%

    No Known Activations