© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50666
    Prev
    Next
    INDEX
    Explanations

    discussing limitations

    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
    è¶ħæłĩ
    -0.11
    åŀĤ
    -0.11
    dess
    -0.09
    åŃĹåı·
    -0.09
    è¿Ŀæ³ķè¿Ŀè§Ħ
    -0.09
    é«ĺæłĩåĩĨ
    -0.08
    åĢĴæĺ¯
    -0.08
    èĬ¯
    -0.08
    å¾ĹçĽĬ
    -0.08
    æİ¥åĬĽ
    -0.08
    POSITIVE LOGITS
    å±ĢéĻIJ
    0.24
     limitations
    0.22
     limitation
    0.19
    limitations
    0.18
     Limit
    0.17
    Limit
    0.16
    éĻIJåĪ¶
    0.16
    缺éĻ·
    0.13
    -limit
    0.13
     limited
    0.13
    Activations Density 0.101%

    No Known Activations