© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 52621
    Prev
    Next
    INDEX
    Explanations

    technical documentation

    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
    Negative Logits
     help
    -0.63
    help
    -0.62
     Help
    -0.52
    帮åĬ©
    -0.52
    	help
    -0.52
    çļĦ帮åĬ©
    -0.49
    (help
    -0.49
    帮
    -0.49
    Help
    -0.48
    _help
    -0.47
    POSITIVE LOGITS
     è¾ĵåĩº
    0.29
    è¾ĵåĩº
    0.28
     output
    0.28
    -Ray
    0.27
     outputs
    0.27
     HOW
    0.26
     наÑħод
    0.26
     Output
    0.25
    /utility
    0.25
     hallway
    0.25
    Activations Density 0.060%

    No Known Activations