© 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. 53201
    Prev
    Next
    INDEX
    Explanations

    short answers and definitions

    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
    --
    -0.36
     co
    -0.31
     c
    -0.31
    ..
    -0.31
    ^
    -0.30
     m
    -0.30
    ...
    -0.30
    e
    -0.30
     &
    -0.30
    <|endoftext|>
    -0.30
    POSITIVE LOGITS
    ’T
    0.30
     scrim
    0.28
    theless
    0.27
     rehearsal
    0.27
     ATTRIBUTE
    0.27
     :↵↵↵↵
    0.27
    stown
    0.27
    SHIP
    0.27
    istrator
    0.26
    itarian
    0.26
    Activations Density 0.044%

    No Known Activations