© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21764
    Prev
    Next
    INDEX
    Explanations

    merger and merge

    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
    ilogy
    -0.09
    ç±į
    -0.09
     обо
    -0.09
    ìĹ´
    -0.08
    大åݦ
    -0.08
    æ³Ħ
    -0.08
    ycz
    -0.08
    å¾Ĺå¾Ī
    -0.08
    nym
    -0.08
     меÑģÑĤо
    -0.08
    POSITIVE LOGITS
    åIJĪ并
    0.20
    èŀįåIJĪ
    0.16
    .Merge
    0.15
     merge
    0.14
     merging
    0.14
    èŀįåħ¥
    0.14
     merges
    0.14
    .merge
    0.13
    merge
    0.13
     merger
    0.13
    Activations Density 0.115%

    No Known Activations