© 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. 20550
    Prev
    Next
    INDEX
    Explanations

    email addresses containing the domain "gmail"

    oai_token-act-pair · gpt-3.5-turbo

    email addresses, particularly those associated with Gmail

    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
    stood
    -0.77
    OTT
    -0.77
    gling
    -0.68
     fitting
    -0.65
     prost
    -0.65
    ©¶æ
    -0.65
    Rated
    -0.63
     corpor
    -0.62
    ŃĶ
    -0.62
    Recomm
    -0.62
    POSITIVE LOGITS
    gmail
    1.25
    yahoo
    0.91
    ileaks
    0.91
    apest
    0.82
    email
    0.81
    legram
    0.71
    zbek
    0.70
    username
    0.70
    allery
    0.69
    agna
    0.68
    Activations Density 0.008%

    No Known Activations