© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 54679
    Prev
    Next
    INDEX
    Explanations

    down

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    åİĿ
    -0.32
     append
    -0.29
     appended
    -0.28
     Almighty
    -0.27
     Append
    -0.26
    į°
    -0.25
    主ä¹īæĢĿæĥ³
    -0.25
     guess
    -0.25
    çĮľ
    -0.25
     guessed
    -0.25
    POSITIVE LOGITS
    -to
    0.27
    slug
    0.26
    si
    0.26
    æ²ī
    0.26
    bugs
    0.26
    -on
    0.24
    GW
    0.24
    WithOptions
    0.24
     spiral
    0.24
    uno
    0.24
    Activations Density 0.008%

    No Known Activations