© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 7124
    Prev
    Next
    INDEX
    Explanations

    instances of the word "engaged" in various contexts such as in relationships, activities, or debates

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "engaged" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oos
    -0.71
    ological
    -0.68
    oland
    -0.63
    ë
    -0.62
    oa
    -0.61
     passage
    -0.61
     Printed
    -0.61
    otropic
    -0.60
    ark
    -0.60
    Origin
    -0.60
    POSITIVE LOGITS
     engaged
    1.01
    ments
    0.94
     engagement
    0.75
    hips
    0.75
    ENC
    0.75
    TAIN
    0.74
    rieved
    0.73
     engages
    0.71
    ENTION
    0.71
    ctl
    0.69
    Activations Density 0.010%

    No Known Activations