© 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. 35-TRANSCODER-HP
    7. 39412
    Prev
    Next
    INDEX
    Explanations

    Different languages/code

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nature
    -0.26
    åĴĮ个人
    -0.26
    ullo
    -0.25
    Heavy
    -0.25
    -tone
    -0.25
     Pang
    -0.25
    izzly
    -0.24
     Heavy
    -0.24
     Morr
    -0.23
    heavy
    -0.23
    POSITIVE LOGITS
    æ
    0.47
    e
    0.45
     e
    0.42
    åĽĽç§į
    0.38
    Óķ
    0.34
    ese
    0.32
    er
    0.32
    ose
    0.31
    æk
    0.31
    ae
    0.31
    Activations Density 65.289%

    No Known Activations