© 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. 19-TRANSCODER-HP
    7. 9138
    Prev
    Next
    INDEX
    Explanations

    Describing, providing or communication

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yer
    -0.25
     dicts
    -0.24
    ehr
    -0.24
    å½ĵçĦ¶
    -0.23
    çĶŁæĢģä¿ĿæĬ¤
    -0.23
    æĥħ
    -0.23
    (custom
    -0.23
    enton
    -0.22
    rels
    -0.22
    æĪij以为
    -0.22
    POSITIVE LOGITS
    æĶĺ
    0.25
    (Equal
    0.25
     Overnight
    0.25
    çŃ
    0.24
    _logs
    0.23
    noc
    0.23
    (compact
    0.23
    ãģĤãĤĬãģ¾ãģĽãĤĵ
    0.23
     accompl
    0.23
     parfait
    0.23
    Activations Density 0.029%

    No Known Activations