© 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. 31-TRANSCODER-HP
    7. 18308
    Prev
    Next
    INDEX
    Explanations

    so

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     till
    -0.49
     hasta
    -0.45
    缴èĩ³
    -0.43
     until
    -0.43
     até
    -0.41
    缴åĪ°
    -0.41
     до
    -0.40
     longer
    -0.39
    åĪ°
    -0.39
     Longer
    -0.37
    POSITIVE LOGITS
     remote
    0.31
     distant
    0.30
     far
    0.30
    è¿ľå¤Ħ
    0.30
    (remote
    0.29
    éģł
    0.28
    è¿ľ
    0.28
    éĻ²
    0.28
    .Remote
    0.28
    è¿ľè¿ľ
    0.28
    Activations Density 0.051%

    No Known Activations