© 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
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51899
    Prev
    Next
    INDEX
    Explanations

    pull back, not thinking, shy away

    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
    apt
    -0.08
    ä¸ĭåİ»
    -0.08
    æįĭ
    -0.08
     UIS
    -0.08
    shake
    -0.08
    omed
    -0.08
     apt
    -0.08
    esti
    -0.08
     nar
    -0.08
    æijĩäºĨ
    -0.08
    POSITIVE LOGITS
    ä¸įæķ¢
    0.16
    å°ıå¿ĥ翼翼
    0.16
    顾èĻij
    0.13
    çĶŁæĢķ
    0.13
    éĥ½ä¸įæķ¢
    0.13
    åħĭåĪ¶
    0.12
    è§Ħéģ¿
    0.12
    æĶ¶æķĽ
    0.12
    é»ĺå¥ij
    0.12
    éģ¿åħį
    0.11
    Activations Density 0.084%

    No Known Activations