© 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. 18335
    Prev
    Next
    INDEX
    Explanations

    claim responsibility or blame

    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
    强迫
    -0.09
    ÑıÑī
    -0.09
    æħķ
    -0.09
    è¾½
    -0.09
    羡
    -0.08
    0
    -0.08
    强åĪ¶
    -0.08
    ived
    -0.08
     projectile
    -0.08
    æĦł
    -0.08
    POSITIVE LOGITS
    躬
    0.13
    åĬŁåĬ³
    0.12
    æī¿æĭħ责任
    0.11
    åĪijäºĭ责任
    0.11
    åĭĩäºİ
    0.11
     responsibility
    0.10
    è´£
    0.10
     GANG
    0.10
    責任
    0.10
    è´Ł
    0.10
    Activations Density 0.028%

    No Known Activations