© 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. 64233
    Prev
    Next
    INDEX
    Explanations

    *

    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
    大åıĶ
    -0.30
     Brands
    -0.30
    å°±è¶Ĭ
    -0.27
     Reserved
    -0.24
    íĶ½
    -0.24
     Moses
    -0.24
     nowhere
    -0.23
     personally
    -0.23
     Sets
    -0.23
    åĴ½åĸī
    -0.23
    POSITIVE LOGITS
     entire
    0.40
     underlying
    0.35
     largest
    0.35
     latest
    0.34
     aforementioned
    0.33
     first
    0.33
     same
    0.33
     slightest
    0.32
     entirety
    0.32
     biggest
    0.31
    Activations Density 93.448%

    No Known Activations