© 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. 14-TRANSCODER-HP
    7. 40038
    Prev
    Next
    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ints
    -0.28
    颧
    -0.27
     wäh
    -0.27
     Seah
    -0.25
    å¼¹æĢ§
    -0.25
    uro
    -0.25
    meye
    -0.25
    ellipsis
    -0.24
    unk
    -0.24
    #else
    -0.24
    POSITIVE LOGITS
    cade
    0.32
     Inject
    0.28
     inject
    0.26
     Injection
    0.25
     injection
    0.24
    uell
    0.23
    éľŃ
    0.23
    sg
    0.23
     injector
    0.23
     fant
    0.23
    Activations Density 1.122%

    No Known Activations