© 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. 56471
    Prev
    Next
    INDEX
    Explanations

    missed or fail

    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
    erce
    -0.10
     vex
    -0.10
    çĿĩ
    -0.09
    ç´¯
    -0.09
     gob
    -0.09
    çĻ½éĵ¶
    -0.09
    çĬĬ
    -0.09
    模ç³Ĭ
    -0.09
    æ©IJ
    -0.09
     Herald
    -0.08
    POSITIVE LOGITS
     missed
    0.17
     misses
    0.17
    miss
    0.15
     miss
    0.15
    åij½ä¸Ń
    0.15
    éĶĻè¿ĩ
    0.14
    _MISS
    0.13
    Miss
    0.13
     Miss
    0.12
    _miss
    0.12
    Activations Density 0.037%

    No Known Activations