© 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. 21290
    Prev
    Next
    INDEX
    Explanations

    to us

    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
    缮åīįåĽ½åĨħ
    -0.10
    -minded
    -0.10
    eus
    -0.09
    aren
    -0.09
    об
    -0.09
    ä»ĸå°±
    -0.09
    çļĦæł·åŃIJ
    -0.09
    typ
    -0.09
     itself
    -0.09
    è¿Ļ家ä¼Ļ
    -0.09
    POSITIVE LOGITS
    phere
    0.13
    ä¿©
    0.13
    /her
    0.13
    ðŁı»
    0.11
    elves
    0.11
    htag
    0.11
    kaar
    0.11
    enet
    0.10
    -même
    0.10
    soever
    0.10
    Activations Density 0.030%

    No Known Activations