© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 36939
    Prev
    Next
    INDEX
    Explanations

    anyone know or have

    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
    饥
    -0.09
    licht
    -0.09
    èĻĶ
    -0.09
    поз
    -0.09
    çľĹ
    -0.09
    å΍
    -0.09
    èĩªç§ģ
    -0.08
    range
    -0.08
    没éĶĻ
    -0.08
    éĩĮæĸ¯
    -0.08
    POSITIVE LOGITS
    è§ģè¿ĩ
    0.10
    xda
    0.10
    çŁ¥éģĵ
    0.10
    hood
    0.10
    çĽ¸ä¼´
    0.09
    çľĭè¿ĩ
    0.08
    .insertBefore
    0.08
     ÑĥÑģÑĤанов
    0.08
    ฯ
    0.08
    说è¿ĩ
    0.08
    Activations Density 0.050%

    No Known Activations