© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51896
    Prev
    Next
    INDEX
    Explanations

    modify it or published by

    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
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    -0.10
    ................................................................
    -0.09
     
    -0.09
    .sourceforge
    -0.09
    enko
    -0.09
    ................................
    -0.08
    æ³Ķ
    -0.08
    .addValue
    -0.08
    ÑĤÑĮ
    -0.08
    ________________________________
    -0.08
    POSITIVE LOGITS
     &↵
    0.17
     \↵
    0.17
    è¶ĬæĿ¥
    0.16
    ï¼Į↵
    0.16
     "+↵
    0.16
    \↵
    0.15
    -↵
    0.14
     [âĢ¦]↵↵
    0.14
    &↵
    0.13
    č↵
    0.13
    Activations Density 3.342%

    No Known Activations