© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 11348
    Prev
    Next
    INDEX
    Explanations

    earned degrees

    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
    ä¹ĭç±»çļĦ
    -0.25
    bol
    -0.23
    ä¹ĭç±»
    -0.23
    æĪĸå°Ĩ
    -0.22
    بÙĪÙĨ
    -0.21
     activity
    -0.21
     general
    -0.20
    æŃ¤ä¸¾
    -0.19
    pire
    -0.19
    ãģ§ãģĻãģĹ
    -0.19
    POSITIVE LOGITS
    è¯ģåĪ¸æĬķèµĦ
    0.24
     spells
    0.22
    åħŃå¹´
    0.22
     Fog
    0.21
    äºĮ次
    0.20
     Spells
    0.20
     subsequ
    0.20
    ikki
    0.19
     spell
    0.19
     Cum
    0.19
    Activations Density 0.088%

    No Known Activations