© 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. 10-RES-JB
    11. 5049
    Prev
    Next
    INDEX
    Explanations

    questions starting with the word "wh-" or "Wh-" followed by other words

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "wh" and its variations, often used in questions or inquiries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mosaic
    -0.67
     curing
    -0.65
     cured
    -0.64
     healing
    -0.63
     BMC
    -0.63
     incub
    -0.62
     fuse
    -0.62
    enegger
    -0.62
     Duo
    -0.62
     Grande
    -0.62
    POSITIVE LOGITS
    ilst
    1.64
    ichever
    1.46
    istle
    1.35
    ats
    1.31
    oles
    1.29
    ole
    1.24
    omever
    1.23
    ispers
    1.22
    soever
    1.21
    olly
    1.14
    Activations Density 0.013%

    No Known Activations