© 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. 38418
    Prev
    Next
    INDEX
    Explanations

    impact and spread

    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
    人å±ħ
    -0.10
    çľī
    -0.10
    ä¼Ľ
    -0.09
    аÑĤÑĥÑĢ
    -0.09
    ä¸İåIJ¦
    -0.09
    çĭ°
    -0.09
    ä¸įçķĻ
    -0.09
    åĮĹ京æŶéĹ´
    -0.08
    ExecutionContext
    -0.08
    以èī²
    -0.08
    POSITIVE LOGITS
    æķĪåºĶ
    0.12
     khắp
    0.12
    è¾IJå°Ħ
    0.12
     ripple
    0.10
    æ·±è¿ľ
    0.09
     effect
    0.09
    å¤ĸåĽ´
    0.09
    è£Ķ
    0.09
     Radiation
    0.09
    éģįåıĬ
    0.09
    Activations Density 0.016%

    No Known Activations