© 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. 30-TRANSCODER-HP
    7. 95459
    Prev
    Next
    INDEX
    Explanations

    conflict and negative events

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    磾
    -0.33
     CREATED
    -0.27
     configurable
    -0.26
    çĶ¸
    -0.26
    çIJ®
    -0.26
     timeStamp
    -0.25
    ices
    -0.25
     realised
    -0.25
     nữa
    -0.25
    eer
    -0.25
    POSITIVE LOGITS
     cost
    0.32
    <Func
    0.30
    æľ¬èº«å°±æĺ¯
    0.29
     would
    0.29
    导
    0.28
    è´¹çĶ¨
    0.28
    enberg
    0.27
     represents
    0.27
     dẫn
    0.26
     mark
    0.26
    Activations Density 0.896%

    No Known Activations