© 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. 19268
    Prev
    Next
    INDEX
    Explanations

    visible

    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
    èļĮ
    -0.24
    urlpatterns
    -0.24
    sweet
    -0.23
    #SBATCH
    -0.23
    enary
    -0.23
     Pent
    -0.23
    quet
    -0.23
    _utilities
    -0.22
    peace
    -0.22
    bw
    -0.22
    POSITIVE LOGITS
    åı¯è§ģ
    1.69
     visible
    1.53
    visible
    1.37
     Visible
    1.34
     vis
    1.31
    Visible
    1.25
     visibility
    1.20
    -visible
    1.16
    vis
    1.12
    (visible
    1.11
    Activations Density 28.870%

    No Known Activations