© 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. 4228
    Prev
    Next
    INDEX
    Explanations

    though followed by contrast

    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
    级以ä¸Ĭ
    -0.12
    å¦Ĥä½ķçľĭå¾ħ
    -0.10
    毫ä¸įçĬ¹è±«
    -0.10
    éļ¾éģĵ
    -0.10
    quota
    -0.10
     indeed
    -0.10
    stor
    -0.09
    å¦Ĥæŀľè¯´
    -0.09
    ropriate
    -0.09
    è¿Ļæł·ä¸Ģ
    -0.09
    POSITIVE LOGITS
    burn
    0.09
    dem
    0.09
    cents
    0.09
    åħ¶
    0.09
    ties
    0.09
    urrent
    0.08
    dh
    0.08
    å®ĥæĺ¯
    0.08
    _ascii
    0.08
    æ·ĭ
    0.08
    Activations Density 0.007%

    No Known Activations