© 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. 54580
    Prev
    Next
    INDEX
    Explanations

    fight or flight

    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
    人人
    -0.11
    åĵĩ
    -0.10
    wel
    -0.10
    èĦį
    -0.09
    大象
    -0.08
    vil
    -0.08
    CNT
    -0.08
    è¿Ļæĸ¹éĿ¢
    -0.08
     Ammo
    -0.08
    Dating
    -0.08
    POSITIVE LOGITS
    æľ¦
    0.10
    æŁĶ
    0.09
    人éĢł
    0.09
    obe
    0.08
    ritable
    0.08
    rede
    0.08
    串
    0.08
     Ind
    0.08
     soft
    0.08
    çī§
    0.08
    Activations Density 0.020%

    No Known Activations