© 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. 10524
    Prev
    Next
    INDEX
    Explanations

    decided to, asked them to, suggested what

    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
    istant
    -0.10
    å®ŀè·µä¸Ń
    -0.10
    èĥĥèĤł
    -0.10
    åķ¡
    -0.09
    ä¸įéĹ´æĸŃ
    -0.09
    ä¸Ģä¸ĩ
    -0.09
    èĥ½æ»¡è¶³
    -0.09
    atching
    -0.09
    /manual
    -0.09
    èķī
    -0.09
    POSITIVE LOGITS
    å¦Ĥä½ķçľĭå¾ħ
    0.10
    mpp
    0.09
    å§Ķå±Ī
    0.09
     rằng
    0.08
    DataStream
    0.08
     поÑĩемÑĥ
    0.08
    çĴ
    0.08
     bahwa
    0.08
    çĶ¨äºº
    0.08
     blah
    0.08
    Activations Density 0.269%

    No Known Activations