© 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. 8117
    Prev
    Next
    INDEX
    Explanations

    witness, witnesses, witnessed

    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
    çģ£
    -0.10
    æī¾äºĨ
    -0.10
    prises
    -0.10
    æĭĴ
    -0.10
    LabelText
    -0.09
    æģ¯
    -0.09
    éĹ°
    -0.09
    ä¸Ģè¡Į
    -0.09
    æ¹¾
    -0.09
    à¹ģà¸Ľ
    -0.09
    POSITIVE LOGITS
    è§ģè¯ģ
    0.27
     witnesses
    0.24
     witness
    0.24
    è§ģè¯ģäºĨ
    0.22
    Witness
    0.21
     witnessing
    0.20
     witnessed
    0.20
    缮çĿ¹
    0.20
    äº²çľ¼
    0.20
     Witness
    0.19
    Activations Density 0.060%

    No Known Activations