© 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. 108764
    Prev
    Next
    INDEX
    Explanations

    Cookies and storage

    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
    åħ¬ç«ł
    -0.28
    mony
    -0.27
    ç¬ĥ
    -0.26
    硬
    -0.25
    лем
    -0.25
     attack
    -0.24
    æĶ¯çIJĥéĺŁ
    -0.24
    åij±
    -0.24
    FullYear
    -0.24
    Sin
    -0.24
    POSITIVE LOGITS
    ắn
    0.26
     Formatting
    0.25
    èį¡
    0.25
    arus
    0.24
     urb
    0.24
     <<
    0.24
    _library
    0.24
    ÅŁim
    0.24
    Formatting
    0.24
    atalog
    0.23
    Activations Density 0.914%

    No Known Activations