© 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. 4238
    Prev
    Next
    INDEX
    Explanations

    kill, death, dying

    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
     TokenType
    -0.09
    ã
    -0.08
    REQUIRED
    -0.08
    ä¸įçŁ¥
    -0.08
    ocale
    -0.08
     jersey
    -0.08
    转移
    -0.08
     justice
    -0.08
    oken
    -0.07
    èĪŀè¹Ī
    -0.07
    POSITIVE LOGITS
    laden
    0.12
    cour
    0.10
     tá»Ń
    0.10
    è¿ĻæĶ¯
    0.10
    _warnings
    0.10
    èĬ¹
    0.10
    èĨº
    0.09
    åħī线
    0.09
    ereal
    0.09
     Grat
    0.08
    Activations Density 0.032%

    No Known Activations