© 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. 53899
    Prev
    Next
    INDEX
    Explanations

    scientific/coding contexts

    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
    å³Ĵ
    -0.30
     commission
    -0.28
     renov
    -0.28
     renaming
    -0.28
    unlock
    -0.28
     modulation
    -0.28
     fixing
    -0.27
     encoding
    -0.27
     fixes
    -0.26
     fixation
    -0.26
    POSITIVE LOGITS
     indexed
    0.59
    indexed
    0.57
     decorated
    0.57
    Indexed
    0.56
     framed
    0.56
    ented
    0.55
     coated
    0.54
    atched
    0.53
    ulated
    0.52
    .logged
    0.52
    Activations Density 0.278%

    No Known Activations