© 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. 19014
    Prev
    Next
    INDEX
    Explanations

    far

    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
    .CG
    -0.32
    ney
    -0.31
    áĥ¥
    -0.28
    igin
    -0.28
     soup
    -0.28
    deaux
    -0.27
    ,long
    -0.26
    ish
    -0.26
    ong
    -0.26
    ungen
    -0.25
    POSITIVE LOGITS
     far
    0.73
    è¿ľ
    0.61
    far
    0.59
    Far
    0.56
    _far
    0.55
     Far
    0.50
    éģł
    0.49
     FAR
    0.46
    è¿ľè¿ľ
    0.45
     reaching
    0.34
    Activations Density 0.142%

    No Known Activations