© 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. 48252
    Prev
    Next
    INDEX
    Explanations

    stranded in water or aboard ships

    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
    äººæł¼
    -0.09
    lix
    -0.09
    é¦ĸéĥ½
    -0.09
    ätt
    -0.08
    antz
    -0.08
     demolished
    -0.08
    ç¥ĸ
    -0.08
    æģ©
    -0.08
     toll
    -0.08
     heav
    -0.08
    POSITIVE LOGITS
    è¢«åĽ°
    0.24
    æ»ŀ
    0.18
     stranded
    0.17
    éĢĥçĶŁ
    0.14
    åĽ°å¢ĥ
    0.13
    åĽ°
    0.13
    åѤ
    0.12
     trapped
    0.12
    èĩªçĦ¶çģ¾å®³
    0.11
    æĹłåĬ©
    0.11
    Activations Density 0.090%

    No Known Activations