© 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. 87705
    Prev
    Next
    INDEX
    Explanations

    scientific texts

    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
     PRI
    -0.28
    èĬĴ
    -0.28
     pr
    -0.28
    就好
    -0.25
    çĽ²
    -0.24
    ity
    -0.24
     Mana
    -0.24
     fasc
    -0.23
    pins
    -0.23
     cement
    -0.23
    POSITIVE LOGITS
    пÑĥÑģ
    0.27
     addButton
    0.24
    _union
    0.24
     PropertyInfo
    0.24
    rina
    0.24
     InputStreamReader
    0.24
     Spicer
    0.23
    allo
    0.23
    AppComponent
    0.23
    icontrol
    0.23
    Activations Density 0.012%

    No Known Activations