© 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. 22-TRANSCODER-HP
    7. 87465
    Prev
    Next
    INDEX
    Explanations

    Code licenses and agreements

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     itself
    -0.28
     Gov
    -0.27
    Gov
    -0.26
    Ëī
    -0.24
     God
    -0.24
     missed
    -0.24
     prom
    -0.24
     ph
    -0.24
    åłµ
    -0.24
     long
    -0.24
    POSITIVE LOGITS
    coh
    0.26
    oral
    0.26
    çī¯
    0.25
     Lau
    0.25
    lexical
    0.25
    amax
    0.24
    elter
    0.24
    .netbeans
    0.23
    heed
    0.23
    _stdio
    0.23
    Activations Density 0.029%

    No Known Activations