© 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. 56393
    Prev
    Next
    INDEX
    Explanations

    file system formatting

    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
    rama
    -0.09
    鲫
    -0.09
    LETTE
    -0.09
    ä¸Ģç»Ħ
    -0.09
    è¤Ĵ
    -0.09
    BUM
    -0.09
    .li
    -0.09
     дог
    -0.09
    AMED
    -0.08
    룹
    -0.08
    POSITIVE LOGITS
    åĪ·
    0.12
    å¼Ł
    0.10
    空çĻ½
    0.09
     chance
    0.09
    åľ°æ¯¯
    0.09
    æ¶Īæ¯Ĵ
    0.09
    好åıĭ
    0.09
    æĻºåķĨ
    0.08
    åį³å°Ĩ
    0.08
     steril
    0.08
    Activations Density 0.080%

    No Known Activations