© 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. 53549
    Prev
    Next
    INDEX
    Explanations

    sensory information processing

    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
    adow
    -0.29
     Hä
    -0.28
     decision
    -0.27
    hus
    -0.26
    decision
    -0.26
    åı¤éķĩ
    -0.25
     proactive
    -0.25
     Ents
    -0.25
    antes
    -0.24
    æIJ¶
    -0.24
    POSITIVE LOGITS
    mess
    0.30
     Detail
    0.28
    odata
    0.28
    mentions
    0.28
     dat
    0.27
    reff
    0.27
    edBy
    0.27
     detail
    0.27
    <Data
    0.27
     ner
    0.27
    Activations Density 0.008%

    No Known Activations