© 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. 19006
    Prev
    Next
    INDEX
    Explanations

    away from

    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
    mma
    -0.10
     drift
    -0.09
    åĶ¯
    -0.08
     Mil
    -0.08
     Lang
    -0.08
    ${
    -0.08
    é¤ħ
    -0.08
    zac
    -0.08
    ç²¥
    -0.08
    æĸ°èĤ¡
    -0.08
    POSITIVE LOGITS
    è¿ľç¦»
    0.13
     away
    0.13
     Away
    0.11
    away
    0.11
    Away
    0.11
    忸
    0.10
    é«ĺæ½®
    0.10
    à¹Ģหà¸Ļ
    0.09
    ÑģÑĥÑĤ
    0.09
    anca
    0.09
    Activations Density 0.007%

    No Known Activations