© 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. 25-TRANSCODER-HP
    7. 108264
    Prev
    Next
    INDEX
    Explanations

    Process reduction and efficiency

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĬĦè¢Ń
    -0.28
     aver
    -0.25
    /events
    -0.25
    è¡ĮæĶ¿æī§æ³ķ
    -0.25
    aclass
    -0.24
     unavoidable
    -0.24
    èĤ¤èī²
    -0.24
    å¯ĤéĿĻ
    -0.24
    moduleId
    -0.24
    opl
    -0.24
    POSITIVE LOGITS
    éĻIJ度
    0.26
    åĬ£åĬ¿
    0.26
    CELL
    0.25
    weg
    0.25
     nisi
    0.25
     tourists
    0.25
     assistance
    0.24
    ccoli
    0.24
    çĩ®
    0.24
    -eff
    0.24
    Activations Density 0.937%

    No Known Activations