© 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. 12048
    Prev
    Next
    INDEX
    Explanations

    forgetting something

    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
    /or
    -0.09
    ä¸įåħ¬å¹³
    -0.09
    IEL
    -0.08
    SEE
    -0.08
    åĿŀ
    -0.08
    loth
    -0.08
     aprove
    -0.08
    ols
    -0.08
    å¥ĩ迹
    -0.08
     thinly
    -0.08
    POSITIVE LOGITS
    çĦ¶æĺ¯
    0.14
    tings
    0.13
    ingly
    0.12
    å¿ĺäºĨ
    0.12
    è¨ĢãģĪãģ°
    0.11
    ting
    0.10
    lob
    0.10
    ëĤł
    0.10
     rằng
    0.10
     lã
    0.09
    Activations Density 0.012%

    No Known Activations