© 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. 25-TRANSCODER-HP
    7. 107890
    Prev
    Next
    INDEX
    Explanations

    computer code

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Corner
    -0.29
     Cabinet
    -0.29
    stry
    -0.27
     cabinet
    -0.27
     cro
    -0.26
    ivid
    -0.25
     gas
    -0.25
     corner
    -0.24
    abinet
    -0.24
    IfExists
    -0.24
    POSITIVE LOGITS
    serve
    0.27
     ç¬
    0.26
     cooperate
    0.25
    atform
    0.25
    æŀ¸
    0.24
    _fact
    0.24
    åĽŀåĽ½
    0.24
     StObject
    0.24
    å°Ĩ士
    0.24
     amused
    0.24
    Activations Density 1.086%

    No Known Activations