© 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. 21615
    Prev
    Next
    INDEX
    Explanations

    somehow

    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
    æĢİä¹ĪåĬŀ
    -0.09
    æīī
    -0.09
    .tail
    -0.08
    åĽł
    -0.08
    èĨı
    -0.08
    为æĤ¨
    -0.08
     wegen
    -0.08
    .ndim
    -0.08
    udas
    -0.08
     heavens
    -0.08
    POSITIVE LOGITS
     somehow
    0.15
    çķª
    0.10
     somew
    0.10
    æŁIJç§į
    0.10
    oodle
    0.09
    æ¶Ľ
    0.09
    åħ¨éķĩ
    0.09
    åıįèĢĮ
    0.09
    说æľį
    0.09
    æ¸ħæ°´
    0.09
    Activations Density 0.069%

    No Known Activations