© 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. 56373
    Prev
    Next
    INDEX
    Explanations

    ending with -ish or -osh

    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
    unda
    -0.09
    èĿī
    -0.08
    -intensive
    -0.08
    undi
    -0.08
    inge
    -0.08
    åĨħåľ¨
    -0.08
    ubb
    -0.08
    alty
    -0.08
    zion
    -0.08
    ittal
    -0.08
    POSITIVE LOGITS
    ä¸ĭ次
    0.10
    磼
    0.10
    产åĮº
    0.09
    ãģıãĤĮ
    0.09
    _FORE
    0.09
    åıįè¿ĩæĿ¥
    0.09
    å½ĵæĪij
    0.09
    åįĥéĩĮ
    0.09
    çħħ
    0.09
    天空
    0.08
    Activations Density 0.063%

    No Known Activations