© 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. 33858
    Prev
    Next
    INDEX
    Explanations

    performing tasks

    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
    æĪIJåĪĨ
    -0.08
     jane
    -0.08
    otic
    -0.08
    é¹ĥ
    -0.08
    åŁĭ
    -0.07
    åľ¨å¤ĸ
    -0.07
    fifo
    -0.07
     bum
    -0.07
    adin
    -0.07
    /stretch
    -0.07
    POSITIVE LOGITS
    ä»»åĬ¡
    0.24
     tasks
    0.22
    tasks
    0.20
     task
    0.18
    ä»»åĭĻ
    0.18
     задаÑĩ
    0.17
     Tasks
    0.17
    Tasks
    0.17
    _tasks
    0.16
    çļĦä»»åĬ¡
    0.15
    Activations Density 0.193%

    No Known Activations