© 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. 30-TRANSCODER-HP
    7. 95498
    Prev
    Next
    INDEX
    Explanations

    historical records and events

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    sp
    -0.29
     Kling
    -0.27
    ä½į
    -0.27
    æ²ģ
    -0.26
    знаком
    -0.25
    ç½ijé¦ĸ页
    -0.25
    atched
    -0.25
    åĨħ容ç®Ģä»ĭ
    -0.25
    éĶ±
    -0.24
    asso
    -0.24
    POSITIVE LOGITS
    yes
    0.26
    _ant
    0.25
    esta
    0.25
    lica
    0.24
    ich
    0.24
     banda
    0.24
    cola
    0.24
    (stdout
    0.23
    fcn
    0.23
    轨
    0.23
    Activations Density 0.883%

    No Known Activations