© 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. 52952
    Prev
    Next
    INDEX
    Explanations

    casting and forging materials

    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
    åĹij
    -0.08
     Hòa
    -0.08
    çļĦçĶ·äºº
    -0.08
     transparent
    -0.08
    éļĶçĥŃ
    -0.08
     prow
    -0.08
    edor
    -0.07
    eca
    -0.07
    äºĽä»Ģä¹Ī
    -0.07
    ÙģÙĨ
    -0.07
    POSITIVE LOGITS
    éĵ¸éĢł
    0.13
    éºĴ
    0.12
    éºŁ
    0.10
    éĽķå¡ij
    0.10
    alar
    0.10
    åħ¬çĽĬæĢ§
    0.10
    äºĨä¸Ģåı£
    0.10
    æ£ł
    0.10
     Casting
    0.10
    è´¤
    0.10
    Activations Density 0.056%

    No Known Activations