© 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. 27-TRANSCODER-HP
    7. 57296
    Prev
    Next
    INDEX
    Explanations

    Long, internet forum texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     exist
    -0.30
     voluntary
    -0.28
    cert
    -0.27
     away
    -0.27
    ächt
    -0.27
     certainty
    -0.26
     fluid
    -0.26
     bi
    -0.26
    篾
    -0.26
     plate
    -0.26
    POSITIVE LOGITS
    Alternatively
    0.33
    Wait
    0.31
    èĥ¸
    0.27
    Qu
    0.26
    istically
    0.25
    wise
    0.25
    leri
    0.25
    Handle
    0.25
    so
    0.25
    ï¼Į↵
    0.25
    Activations Density 0.001%

    No Known Activations