© 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. 54406
    Prev
    Next
    INDEX
    Explanations

    Taking something into account

    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
    /story
    -0.27
    é¥Ń
    -0.26
    —who
    -0.25
     bestellen
    -0.25
    retrieve
    -0.25
    rq
    -0.24
    å¨ĺå¨ĺ
    -0.24
    åĶ¤
    -0.24
    -tip
    -0.24
    -Smith
    -0.23
    POSITIVE LOGITS
     due
    0.59
    due
    0.46
     under
    0.45
    åĽł
    0.43
     Due
    0.43
    Due
    0.41
     int
    0.40
    çĶ±äºİ
    0.40
    _due
    0.36
    under
    0.34
    Activations Density 0.015%

    No Known Activations