© 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. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 141882
    Prev
    Next
    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĵºè®¾
    -0.28
    ekt
    -0.26
    Âł
    -0.26
    ebb
    -0.25
    ä¾µæĿĥ
    -0.25
     Creative
    -0.25
     Innovative
    -0.25
    æĪĸ许
    -0.24
    è¿Ŀ纪
    -0.24
    äºĮ线
    -0.23
    POSITIVE LOGITS
    åŁºåĩĨ
    0.30
    个çĻ¾åĪĨ
    0.27
     Mourinho
    0.26
    èıĮ
    0.25
     stakes
    0.25
    èĿĩ
    0.25
     ($)
    0.24
     Garrett
    0.24
     autogenerated
    0.24
    ä¹ĭäºī
    0.24
    Activations Density 0.389%

    No Known Activations