© 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. 18670
    Prev
    Next
    INDEX
    Explanations

    code and links

    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
    è¿ĶåĽŀ
    -0.48
     return
    -0.40
     Return
    -0.37
    è¿Ķ
    -0.37
     returning
    -0.35
    .Return
    -0.35
    returns
    -0.34
     è¿ĶåĽŀ
    -0.34
     returns
    -0.34
    return
    -0.34
    POSITIVE LOGITS
    intr
    0.29
    endet
    0.27
    croft
    0.26
    staff
    0.26
    ẵ
    0.25
     Closing
    0.25
    borough
    0.25
     Intr
    0.24
    à¤ĺ
    0.24
    osph
    0.24
    Activations Density 0.002%

    No Known Activations