© 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. 39809
    Prev
    Next
    INDEX
    Explanations

    researchers/team discovering/reporting

    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
    acas
    -0.09
    seen
    -0.08
    åıijè¾¾åĽ½å®¶
    -0.08
    æĭĴ
    -0.08
    ç¥ŀå·ŀ
    -0.08
    被æī§è¡Į
    -0.08
    æĪij们éĥ½
    -0.08
    unik
    -0.08
    让ä»ĸ们
    -0.08
    /fs
    -0.08
    POSITIVE LOGITS
    ãĥģãĥ¼ãĥł
    0.09
    åIJĮåŁİ
    0.09
    èݹ
    0.09
    åı¯çĪ±çļĦ
    0.09
    å°ıå§IJå§IJ
    0.09
    acker
    0.09
    è¿ľæĸ¹
    0.08
    åĽ¢éĺŁ
    0.08
    æ±ŁåĮĹ
    0.08
    å¸ĮæľĽèĥ½å¤Ł
    0.08
    Activations Density 0.079%

    No Known Activations