© 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. 20-TRANSCODER-HP
    7. 107234
    Prev
    Next
    INDEX
    Explanations

    and, comma

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Oc
    -0.26
    -local
    -0.26
     nomination
    -0.26
    akash
    -0.26
    æĺ¯éŁ©åĽ½
    -0.25
    éĴ«
    -0.23
    izar
    -0.23
    ستاÙĨ
    -0.23
    .Output
    -0.23
    hood
    -0.23
    POSITIVE LOGITS
    æī°
    0.31
    /articles
    0.27
    å¼
    0.26
    aments
    0.25
     contacts
    0.25
    iera
    0.25
    /compiler
    0.24
    mel
    0.24
     shl
    0.24
    isible
    0.24
    Activations Density 33.415%

    No Known Activations