© 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. 31-TRANSCODER-HP
    7. 18802
    Prev
    Next
    INDEX
    Explanations

    political names/events

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ed
    -0.40
    edb
    -0.37
    edx
    -0.36
    eds
    -0.34
    edException
    -0.31
     bust
    -0.31
    edm
    -0.29
    ted
    -0.29
     ed
    -0.29
    -ed
    -0.28
    POSITIVE LOGITS
     Had
    0.40
    Had
    0.38
    çĶ¨äºĨ
    0.37
    had
    0.33
     prepares
    0.32
     byÅĤa
    0.32
    åıijæĮ¥äºĨ
    0.32
     Preparation
    0.32
     preparing
    0.30
    .prepare
    0.30
    Activations Density 1.778%

    No Known Activations