© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20747
    Prev
    Next
    INDEX
    Explanations

    failure or collapse

    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
    çĺ³
    -0.10
    鬼
    -0.09
    adox
    -0.08
    lox
    -0.08
    éļ¶å±ŀäºİ
    -0.08
    碼
    -0.08
    çłģ
    -0.08
    _inverse
    -0.08
     repar
    -0.08
    agon
    -0.08
    POSITIVE LOGITS
    失败
    0.17
     failure
    0.16
    failure
    0.15
    失æķĪ
    0.14
     failures
    0.14
     failed
    0.14
     Failure
    0.14
     Fail
    0.14
     fail
    0.13
     FAILURE
    0.13
    Activations Density 0.093%

    No Known Activations