© 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. 1896
    Prev
    Next
    INDEX
    Explanations

    copyright, author, email

    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
    -Agent
    -0.10
    CLAIM
    -0.09
    åĵ¨
    -0.09
     Sponge
    -0.09
    ADER
    -0.08
     tuy
    -0.08
    adesh
    -0.08
    æĿ¥è¿ĻéĩĮ
    -0.08
    akk
    -0.08
    launcher
    -0.08
    POSITIVE LOGITS
    çĪ¶åŃIJ
    0.09
    æĺ¯ä»ĸ
    0.09
    otics
    0.08
    rovers
    0.08
    ready
    0.08
    esser
    0.08
     Ready
    0.08
     reserved
    0.08
    rescia
    0.08
     reserve
    0.08
    Activations Density 0.198%

    No Known Activations