© 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. 35-TRANSCODER-HP
    7. 40331
    Prev
    Next
    INDEX
    Explanations

    scientific descriptions and inventions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     stret
    -0.26
     thả
    -0.26
     squeez
    -0.25
     pudd
    -0.25
     altern
    -0.25
     progressively
    -0.25
     extingu
    -0.25
     accomp
    -0.25
     NHS
    -0.24
    èħĮ
    -0.24
    POSITIVE LOGITS
    ional
    0.39
    itions
    0.36
    isions
    0.35
    iations
    0.34
    rical
    0.34
    icians
    0.33
    imeter
    0.33
    rics
    0.33
    iros
    0.33
    ters
    0.32
    Activations Density 0.034%

    No Known Activations