© 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. 53603
    Prev
    Next
    INDEX
    Explanations

    measurement units

    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
    æİĸ
    -0.28
    å®īæ£Ģ
    -0.28
    å«ģç»Ļ
    -0.26
     Inhal
    -0.25
    adge
    -0.25
     indifferent
    -0.25
    stations
    -0.24
    ortality
    -0.24
    DROP
    -0.24
    hma
    -0.24
    POSITIVE LOGITS
     on
    0.33
    å±Ģåī¯å±Ģéķ¿
    0.27
     unit
    0.26
    _unit
    0.26
    夫
    0.26
    'on
    0.26
    è´Ł
    0.26
     dig
    0.26
    unit
    0.25
     onStart
    0.25
    Activations Density 0.062%

    No Known Activations