© 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. 0-RES-JB
    11. 10324
    Prev
    Next
    INDEX
    Explanations

    names, particularly the name "Brent"

    oai_token-act-pair · gpt-3.5-turbo

    names of individuals, particularly those similar or related to "Brent" and "Robertson"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    é¾į
    -0.75
     selfie
    -0.75
     selfies
    -0.73
    xual
    -0.71
     unicorn
    -0.70
    Í
    -0.69
     ninja
    -0.68
     bikes
    -0.67
     healer
    -0.67
    ktop
    -0.67
    POSITIVE LOGITS
     Brent
    2.36
    ei
    1.76
     Robertson
    1.45
     Schwartz
    1.14
     Stern
    1.10
     Banner
    1.10
     Weir
    1.09
     Gross
    1.05
     Barnett
    1.03
    Rating
    0.97
    Activations Density 0.038%

    No Known Activations