© 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. 39161
    Prev
    Next
    INDEX
    Explanations

    pattern of behavior, instruction, animation, measurement, use

    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
    端æŃ£
    -0.09
     LIMITED
    -0.08
    å¤ıåŃ£
    -0.08
    éģĩåĪ°äºĨ
    -0.08
    éĿ¢åŃIJ
    -0.08
     Limited
    -0.08
    æ±Ľ
    -0.08
     огÑĢаниÑĩен
    -0.08
     regularization
    -0.08
     specialised
    -0.08
    POSITIVE LOGITS
     action
    0.11
    æĸ§
    0.10
     work
    0.10
     support
    0.09
    çĥŃéĩı
    0.09
    è¡ĮåĬ¨
    0.09
     understanding
    0.09
     functionality
    0.09
    -operation
    0.08
    æĢĿèĢĥ
    0.08
    Activations Density 0.168%

    No Known Activations