© 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. 33-TRANSCODER-HP
    7. 54048
    Prev
    Next
    INDEX
    Explanations

    ens/Sens

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    horn
    -0.33
    çĶŁäº§æĢ»åĢ¼
    -0.30
    orn
    -0.28
    berries
    -0.28
    (URL
    -0.28
     ditch
    -0.27
     excit
    -0.27
     horn
    -0.26
     regress
    -0.25
    ç§ģæľį
    -0.25
    POSITIVE LOGITS
    e
    0.28
    iage
    0.27
    i
    0.26
    าว
    0.25
    æĿ·
    0.24
    ee
    0.24
    齿
    0.24
    ec
    0.24
    è¾¾åĪ°
    0.23
    Pin
    0.23
    Activations Density 0.030%

    No Known Activations