© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 1395
    Prev
    Next
    INDEX
    Explanations

    #endif or code block end

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
     tre
    -0.23
    \xa
    -0.22
     blind
    -0.22
     bow
    -0.22
    \u
    -0.22
     Cou
    -0.21
     Revolution
    -0.21
    odem
    -0.21
     turn
    -0.21
     ti
    -0.21
    POSITIVE LOGITS
    çŃīçŃī
    0.34
    etc
    0.32
    åĩłåı¥
    0.30
     blah
    0.28
    å¾Ģä¸ĭ
    0.28
     etc
    0.27
    ä¸Ģ大åłĨ
    0.27
    -placeholder
    0.27
    ä¹ĭç±»çļĦ
    0.26
    ä¸įè§ģäºĨ
    0.26
    Activations Density 0.139%

    No Known Activations