© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21003
    Prev
    Next
    INDEX
    Explanations

    witnesses and interviews

    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
    prises
    -0.10
    说æĪij
    -0.09
    charge
    -0.09
    è°Ģ
    -0.09
    eling
    -0.09
    alist
    -0.09
    AGER
    -0.08
    åīĶ
    -0.08
    æĪijä¸Ģ缴
    -0.08
    .asList
    -0.08
    POSITIVE LOGITS
    åıĸè¯ģ
    0.11
    磮
    0.10
    è¯ģ
    0.10
    çĨŁ
    0.10
     witnesses
    0.10
    Interview
    0.10
    åıĹ害
    0.10
     interview
    0.09
    走访
    0.09
    约è°Ī
    0.09
    Activations Density 0.054%

    No Known Activations