© 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. 39858
    Prev
    Next
    INDEX
    Explanations

    technical/code related

    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
     vid
    -0.26
     jams
    -0.25
     clear
    -0.25
    ç»ıéªĮ丰å¯Į
    -0.24
     habitat
    -0.24
     alternative
    -0.24
    aman
    -0.24
     пÑĢоÑĤ
    -0.24
     adjud
    -0.23
     shred
    -0.23
    POSITIVE LOGITS
    lin
    0.30
     WOW
    0.29
    lein
    0.28
    oron
    0.28
    ledge
    0.27
    lings
    0.27
    l
    0.26
    ling
    0.26
    letal
    0.26
    UPPORT
    0.26
    Activations Density 0.189%

    No Known Activations