© 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. 47904
    Prev
    Next
    INDEX
    Explanations

    odds and ends

    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
    ungs
    -0.09
    æĭĵå±ķ
    -0.09
    ned
    -0.09
     Ventura
    -0.08
    æľ¬åľŁ
    -0.08
     Äijẹ
    -0.08
    åĪĨå·¥
    -0.08
     PROFITS
    -0.08
    æĬij
    -0.08
    æľīä¸Ģå®ļçļĦ
    -0.08
    POSITIVE LOGITS
    Bits
    0.12
    æĭ¾
    0.12
    bits
    0.10
    好è¯Ħ
    0.09
    æ³Ħéľ²
    0.09
    åIJĦç§į
    0.09
     off
    0.09
    åIJĥè¿ĩ
    0.09
     bits
    0.09
     Bits
    0.09
    Activations Density 0.029%

    No Known Activations