© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 4-TRANSCODER-HP
    7. 215
    Prev
    Next
    INDEX
    Explanations

    iz

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_4.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.4.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     elim
    -0.26
    di
    -0.25
    Tho
    -0.24
    Anth
    -0.24
    磼
    -0.23
    æĺ§
    -0.23
    igma
    -0.23
     ami
    -0.23
     Anth
    -0.23
     answered
    -0.23
    POSITIVE LOGITS
    ousel
    0.26
    ILER
    0.26
    ajas
    0.26
     kê
    0.25
    .jquery
    0.25
    uju
    0.25
    ilig
    0.24
    说æľį
    0.24
    .nick
    0.24
    ullah
    0.23
    Activations Density 0.016%

    No Known Activations