© 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. 18-TRANSCODER-HP
    7. 89224
    Prev
    Next
    INDEX
    Explanations

    war

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å°¼
    -0.30
    ä¸į容æĺĵ
    -0.26
    bestos
    -0.26
    oft
    -0.25
    beck
    -0.25
    å³Ļ
    -0.24
    æľīä¸Ģç§į
    -0.24
    aż
    -0.24
    bbie
    -0.24
    ored
    -0.23
    POSITIVE LOGITS
    ONSE
    0.28
    课å¤ĸ
    0.27
    unchecked
    0.26
    Incomplete
    0.26
     inplace
    0.26
    è¿Ľå±ķæĥħåĨµ
    0.25
     $($
    0.25
    èįļ
    0.25
    å¯ĨåĪĩ
    0.24
    cut
    0.24
    Activations Density 18.119%

    No Known Activations