© 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. 29908
    Prev
    Next
    INDEX
    Explanations

    shaped or shaping influence

    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
    è¶Ĭé«ĺ
    -0.10
    aben
    -0.09
    éĽ¢ãĤĮ
    -0.09
    çĽij
    -0.08
    ylon
    -0.08
    gment
    -0.08
    è·ĥ
    -0.08
    REEN
    -0.08
    éªĩ
    -0.08
    yc
    -0.08
    POSITIVE LOGITS
     shaping
    0.13
     shape
    0.11
     history
    0.11
    çļĦå½±åĵį
    0.10
     Cle
    0.10
    æĹ©æĹ©
    0.09
    å½¢çĬ¶
    0.09
    æīĵéĢłåĩº
    0.09
    æ¾¹
    0.09
     Influ
    0.09
    Activations Density 0.049%

    No Known Activations