© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 12636
    Prev
    Next
    INDEX
    Explanations

    balance and stability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    大è§Ħ模
    -0.21
     Cres
    -0.20
    éļIJèĶ½
    -0.20
    大æī¹
    -0.20
    亲åĪĩ
    -0.19
    ired
    -0.19
     curtain
    -0.19
    å¼ķãģ£
    -0.19
    èĨı
    -0.19
    çĸĻ
    -0.19
    POSITIVE LOGITS
    平衡
    0.39
    稳å®ļæĢ§
    0.36
    ä¸į稳å®ļ
    0.35
     unstable
    0.35
     balance
    0.34
    稳å®ļ
    0.34
     stability
    0.34
     precarious
    0.33
    (balance
    0.31
    ãĥIJãĥ©ãĥ³ãĤ¹
    0.31
    Activations Density 0.050%

    No Known Activations