© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19487
    Prev
    Next
    INDEX
    Explanations

    clean

    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
    èĩ¾
    -0.10
    itim
    -0.10
    åıĺå¹»
    -0.09
    amble
    -0.09
    ç¾İ人
    -0.09
    elve
    -0.09
    alom
    -0.09
    поÑĤÑĢеб
    -0.08
    åijĹ
    -0.08
    itti
    -0.08
    POSITIVE LOGITS
    å¹²åĩĢ
    0.15
    liness
    0.14
    est
    0.12
    æķ´æ´ģ
    0.11
    濯
    0.11
    -minded
    0.10
     minded
    0.10
    æ´ģåĩĢ
    0.09
    èĮ¹
    0.09
    æī«ä¸Ģ
    0.09
    Activations Density 0.034%

    No Known Activations