© 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. 21442
    Prev
    Next
    INDEX
    Explanations

    conflict and difficult interactions

    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
     Wass
    -0.09
    CRYPT
    -0.09
    å¹ķ
    -0.09
    ucer
    -0.09
    inges
    -0.09
    ç«
    -0.09
     Warn
    -0.08
    å¼ĥ
    -0.08
    çĹĽ
    -0.08
    ToInt
    -0.08
    POSITIVE LOGITS
    é£İæł¼
    0.12
    åĨ²çªģ
    0.12
    åIJµæŀ¶
    0.11
    éĿ¢å¯¹
    0.10
     Styles
    0.10
    äºīåIJµ
    0.10
     confrontation
    0.10
    æģ¼
    0.10
     styles
    0.09
    Conflict
    0.09
    Activations Density 0.117%

    No Known Activations