© 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. 63594
    Prev
    Next
    INDEX
    Explanations

    consequences 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
    æŃ¤åŁºç¡Ģä¸Ĭ
    -0.11
    rock
    -0.10
    èĻ±
    -0.10
    worm
    -0.10
    ahi
    -0.10
    笳
    -0.09
    èĪ«
    -0.09
    æĹĮ
    -0.09
    çŀij
    -0.09
    åĻ
    -0.09
    POSITIVE LOGITS
    ental
    0.10
    æ·±è¿ľ
    0.09
    ality
    0.09
    nel
    0.09
    Ãłnh
    0.09
    iq
    0.09
    ively
    0.08
    SPATH
    0.08
    ãĤį
    0.08
    ADO
    0.08
    Activations Density 0.015%

    No Known Activations