© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51343
    Prev
    Next
    INDEX
    Explanations

    costs, losses, and recovery

    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
     PROPERTY
    -0.09
    ailable
    -0.09
    rtype
    -0.08
    éĽħ
    -0.08
    éĽĨ
    -0.08
    -headed
    -0.08
    å®Ī
    -0.08
    å¿Ļ
    -0.08
    æĶ¹
    -0.08
     Gar
    -0.08
    POSITIVE LOGITS
    æĶ¶åĽŀ
    0.14
    æijĬ
    0.13
    åĦŁ
    0.11
    æĪIJæľ¬
    0.11
    charge
    0.11
    ä¸įè´Ł
    0.10
    弥补
    0.10
     Ñģеб
    0.10
    æ±²åıĸ
    0.10
    æĤī
    0.10
    Activations Density 0.053%

    No Known Activations