© 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. 58297
    Prev
    Next
    INDEX
    Explanations

    suicide

    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
    éģ²
    -0.10
    ä¸įè¡Į
    -0.09
    å¾·å°Ķ
    -0.09
    .vol
    -0.09
    viz
    -0.09
    criptors
    -0.09
    çķª
    -0.08
    çļĦåĪ©çĽĬ
    -0.08
    èĩŁ
    -0.08
    rray
    -0.08
    POSITIVE LOGITS
    èĩªæĿĢ
    0.28
     suicide
    0.19
     suicides
    0.18
     suic
    0.17
     suicidal
    0.14
    æĬijéĥģçĹĩ
    0.13
     Suicide
    0.13
    ç¼¢
    0.11
    æĬijéĥģ
    0.11
     antidepress
    0.11
    Activations Density 0.147%

    No Known Activations