© 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. 52970
    Prev
    Next
    INDEX
    Explanations

    High/low pressure, power, temperature

    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
    irc
    -0.26
    жен
    -0.25
    ãĤ«ãĥ³
    -0.24
    pike
    -0.24
    itta
    -0.24
     imperative
    -0.24
    wcs
    -0.24
     communal
    -0.23
    ells
    -0.23
    ikk
    -0.23
    POSITIVE LOGITS
     flex
    0.38
     tor
    0.33
    çĤ¹è¯Ħ
    0.27
    end
    0.27
    Tor
    0.27
    tor
    0.27
    flex
    0.26
    -output
    0.26
    mer
    0.26
    	flex
    0.26
    Activations Density 0.018%

    No Known Activations