© 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. 33-TRANSCODER-HP
    7. 53609
    Prev
    Next
    INDEX
    Explanations

    moving away

    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
    Not in Any Lists

    No Comments

    Negative Logits
    çľīæ¯Ľ
    -0.26
    çľī头
    -0.26
    æ³½
    -0.25
    åĨ¢
    -0.25
    åıĮèħ¿
    -0.25
    ÅĪ
    -0.24
    Bird
    -0.24
    è¿Ļç§įæĸ¹æ³ķ
    -0.24
    зеÑĢ
    -0.23
     encoding
    -0.23
    POSITIVE LOGITS
     way
    0.40
     clear
    0.36
     ar
    0.32
     а
    0.30
     Clear
    0.30
     clears
    0.30
    .clear
    0.29
     Way
    0.28
     ways
    0.28
     an
    0.27
    Activations Density 0.016%

    No Known Activations