© 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. 6694
    Prev
    Next
    INDEX
    Explanations

    expand, expanded, expansion

    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
    æĺ§
    -0.11
    çĤ¬
    -0.11
    CharCode
    -0.10
    è§Ĩ线
    -0.09
    éħ¿
    -0.09
    çģı
    -0.09
    è¿Ľåľº
    -0.09
    éģIJ
    -0.09
    furt
    -0.09
    éĢĨè¡Į
    -0.09
    POSITIVE LOGITS
    èĨ¨èĥĢ
    0.19
    æī©å¼ł
    0.15
     Expand
    0.13
     Expansion
    0.13
    èĨ¨
    0.13
     expansion
    0.13
    Expand
    0.12
     expanded
    0.12
    Expansion
    0.12
    èĥĢ
    0.12
    Activations Density 0.155%

    No Known Activations