© 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. 5310
    Prev
    Next
    INDEX
    Explanations

    resources and recovery

    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
    æķ¬
    -0.12
    ä¸įæĺ¯ä¸Ģ个
    -0.11
    åij¨è½¬
    -0.10
    æ¯ĽåŃĶ
    -0.10
    佩æľį
    -0.10
    èĩªåĪ¶
    -0.10
    ä¸Ģå¹ķ
    -0.09
    åľ¨å¸Ĥåľºä¸Ĭ
    -0.09
    éĺĶ
    -0.09
    illet
    -0.09
    POSITIVE LOGITS
    æľªç»ı
    0.10
    èķ´
    0.10
    丰åİļ
    0.09
    ç¦Ģ
    0.09
     tutor
    0.08
    å¤©åºľ
    0.08
    entina
    0.08
     unt
    0.08
     lucrative
    0.08
    ä¹Łåıªèĥ½
    0.08
    Activations Density 0.084%

    No Known Activations