© 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. 1427
    Prev
    Next
    INDEX
    Explanations

    make errors

    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
    è°
    -0.10
     Moody
    -0.09
    anged
    -0.09
     Decomp
    -0.09
    é«ĺæĸ°æĬĢæľ¯
    -0.09
    arrow
    -0.09
    ivité
    -0.09
    è±ģ
    -0.08
    çĻ¾å®¶
    -0.08
    é«ij
    -0.08
    POSITIVE LOGITS
     errors
    0.16
     error
    0.16
    误差
    0.15
     mistakes
    0.15
    errors
    0.14
    error
    0.14
     оÑĪиб
    0.14
    çºłéĶĻ
    0.14
    çĬ¯
    0.14
    éĶĻ误
    0.14
    Activations Density 0.116%

    No Known Activations