© 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. 63383
    Prev
    Next
    INDEX
    Explanations

    skill acquisition and training

    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
    主æµģ
    -0.09
     apprec
    -0.09
    ç»ıè´¸
    -0.08
    åīįéĶĭ
    -0.08
    _IDLE
    -0.08
    é¬ĵ
    -0.08
    heading
    -0.08
    椽
    -0.08
    ipse
    -0.08
     Apprec
    -0.08
    POSITIVE LOGITS
    åŁ¹è®Ń
    0.13
    æķijæĬ¤
    0.13
    èĢĥæł¸
    0.11
    ä¸Ģéģį
    0.11
    æĢ¥æķij
    0.11
    å²Ĺ
    0.10
    ä¸Ĭå²Ĺ
    0.10
    æĢĢæĬ±
    0.10
    æ¼Ķä¹ł
    0.10
     training
    0.09
    Activations Density 0.110%

    No Known Activations