© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 1-RES-JB
    11. 20400
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically the name "Kent"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Kent."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    thood
    -0.89
    llah
    -0.78
    xual
    -0.77
    BOOK
    -0.76
    units
    -0.73
    eworld
    -0.70
    estyles
    -0.70
    senal
    -0.70
    verbal
    -0.69
    ework
    -0.69
    POSITIVE LOGITS
     Kent
    1.28
     Dover
    0.90
    rell
    0.85
    alus
    0.83
    Kent
    0.81
     Somers
    0.80
     Sussex
    0.79
     Walton
    0.77
     Baz
    0.76
     Barton
    0.76
    Activations Density 0.003%

    No Known Activations