© 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. 49909
    Prev
    Next
    INDEX
    Explanations

    channeling or directing

    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
    avig
    -0.09
    од
    -0.09
    延误
    -0.09
    è¡Ģ液循çݯ
    -0.08
    ди
    -0.08
    nton
    -0.08
    ода
    -0.08
    çĩİ
    -0.08
    çŀŃ
    -0.08
    ensus
    -0.08
    POSITIVE LOGITS
    èĢĮæĪIJ
    0.10
    inar
    0.10
     cer
    0.09
    å¿ĥæĢĿ
    0.09
    ç²¾åĬĽ
    0.09
    ties
    0.09
    åħ½
    0.09
    åĨħåľ°
    0.09
    åŁ¤
    0.09
    wor
    0.09
    Activations Density 0.004%

    No Known Activations