© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 2476
    Prev
    Next
    INDEX
    Explanations

    click image to enlarge

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
    ùng
    -0.24
    éģ¿
    -0.24
    TypeInfo
    -0.22
    .normalize
    -0.21
    Wire
    -0.21
     stere
    -0.21
    omens
    -0.20
    VertexBuffer
    -0.20
    speed
    -0.20
     bell
    -0.20
    POSITIVE LOGITS
    ä¸ĭæĸ¹
    0.31
     pictured
    0.29
    ä¸Ĭæĸ¹
    0.28
    请çĤ¹åĩ»
    0.26
     BELOW
    0.25
    wood
    0.24
    lein
    0.24
    Ả
    0.24
    éĴľ
    0.24
    bove
    0.23
    Activations Density 0.010%

    No Known Activations