© 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. 39457
    Prev
    Next
    INDEX
    Explanations

    theft over Auto

    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
    keit
    -0.11
    meer
    -0.10
     nowhere
    -0.10
     оÑĤдÑĭÑħ
    -0.10
    íŀĪ
    -0.10
    ität
    -0.10
    å°¼äºļ
    -0.09
    æķĻè®Ń
    -0.09
    ationToken
    -0.09
    ForResult
    -0.08
    POSITIVE LOGITS
    entimes
    0.10
    rets
    0.09
    ายà¸Ļ
    0.09
    沫
    0.09
    eter
    0.09
    ened
    0.08
    èĩªè´Ł
    0.08
    ancock
    0.08
    rix
    0.08
    è¡ĹåĮº
    0.08
    Activations Density 0.005%

    No Known Activations