© 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. 39686
    Prev
    Next
    INDEX
    Explanations

    wrong input, subsequent attempts

    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
    èĢĮéĿŀ
    -0.10
    æ´Ĺ
    -0.09
    åī¯ä¸»ä»»
    -0.09
    abant
    -0.09
    dz
    -0.08
    acco
    -0.08
    ORK
    -0.08
    ä¸Ģ缴åĪ°
    -0.08
    ëĵ
    -0.08
     quán
    -0.08
    POSITIVE LOGITS
    æĢİä¹ĪåĬŀ
    0.14
    该æĢİä¹ĪåĬŀ
    0.11
    ä¹Łä¸įä¼ļ
    0.10
    è¦ģåıĬæŶ
    0.10
    æĻĤãģ«
    0.10
    åıĬæŶ
    0.10
    ä½Ĩä»į
    0.10
    æħ°
    0.10
    asan
    0.10
    çĽĬ
    0.10
    Activations Density 0.169%

    No Known Activations