© 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. 18411
    Prev
    Next
    INDEX
    Explanations

    forensic and investigative details

    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
    -bodied
    -0.11
    į
    -0.09
    à¹Ģà¸Ħล
    -0.09
    ın
    -0.09
    uir
    -0.09
    oire
    -0.09
    oral
    -0.08
    -Language
    -0.08
     Perr
    -0.08
    ä»·éĴ±
    -0.08
    POSITIVE LOGITS
    å·¥ä½ľèĢħ
    0.12
    åŃ¦å®¶
    0.10
    -minded
    0.10
     här
    0.10
    å¾ĹåĪ°æľīæķĪ
    0.09
    è´º
    0.09
    Hub
    0.09
    ×Ļ×ķת
    0.09
    çĻĸ
    0.09
    è¿ĩå¾Ĺ
    0.09
    Activations Density 0.006%

    No Known Activations