© 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. 17243
    Prev
    Next
    INDEX
    Explanations

    further X

    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
    sWith
    -0.10
    æıĨ
    -0.09
    ÑĤ
    -0.09
    éļıå¤Ħ
    -0.09
    éĥ½è¯´
    -0.09
    åIJİèĢħ
    -0.09
    ино
    -0.09
    æİĴè¡Įæ¦ľ
    -0.09
    pragma
    -0.09
    åīįèĢħ
    -0.09
    POSITIVE LOGITS
    -reaching
    0.12
     ado
    0.12
    ance
    0.12
    hin
    0.11
    most
    0.11
    mente
    0.10
    ment
    0.10
    -than
    0.10
    CTION
    0.10
    à¹Į
    0.10
    Activations Density 0.025%

    No Known Activations