© 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. 107975
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    ricks
    -0.26
    rid
    -0.25
     RID
    -0.25
    affer
    -0.25
    天使
    -0.25
    ["$
    -0.24
     Defaults
    -0.24
    adas
    -0.24
     pain
    -0.24
    seconds
    -0.24
    POSITIVE LOGITS
    å£ħ
    0.26
    Compound
    0.26
     Compound
    0.25
     beh
    0.25
    á»įc
    0.24
     İç
    0.24
     schle
    0.24
     compound
    0.23
    ÃĵN
    0.23
    è¹Ī
    0.23
    Activations Density 0.160%

    No Known Activations

    This feature has no known activations.