© 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. 10342
    Prev
    Next
    INDEX
    Explanations

    mentions of the author "By"

    oai_token-act-pair · gpt-3.5-turbo

    the word "By" followed by names or authors

    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
    itives
    -0.72
    Finish
    -0.71
    ilib
    -0.65
    vation
    -0.64
    abs
    -0.64
    ickets
    -0.61
    inished
    -0.61
    ilogy
    -0.60
    icken
    -0.58
    medi
    -0.58
    POSITIVE LOGITS
    akuya
    0.95
     DAV
    0.91
     Brett
    0.90
     Stephanie
    0.88
     Dustin
    0.88
     Ellen
    0.87
     Patrick
    0.87
     JUL
    0.87
     Jonathan
    0.87
     Judith
    0.87
    Activations Density 0.022%

    No Known Activations