© 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. 1781
    Prev
    Next
    INDEX
    Explanations

    invalid email addresses

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Email."

    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
     prim
    -0.74
     stable
    -0.67
     Nob
    -0.65
     par
    -0.64
     stereotyp
    -0.63
     survival
    -0.63
     recognized
    -0.63
     plurality
    -0.62
     nobility
    -0.62
     scen
    -0.62
    POSITIVE LOGITS
     Email
    3.76
    Email
    2.39
     Emails
    2.19
     email
    2.18
    email
    1.94
     emails
    1.72
    mail
    1.58
    Mail
    1.47
     emailed
    1.43
     Gmail
    1.41
    Activations Density 0.010%

    No Known Activations