© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 2-RES-JB
    11. 14139
    Prev
    Next
    INDEX
    Explanations

    similarities or patterns in references to specific names or identifiers, particularly "Doe" and "Kurd"

    oai_token-act-pair · gpt-3.5-turbo

    references to legal cases involving individuals identified as "Jane Doe."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rams
    -0.80
    cius
    -0.76
    ropolitan
    -0.74
    rist
    -0.74
    arching
    -0.73
    ories
    -0.71
    eways
    -0.69
    OIL
    -0.69
    arta
    -0.69
    Capital
    -0.69
    POSITIVE LOGITS
     Doe
    1.28
    ele
    0.74
    zie
    0.73
     accuser
    0.71
    ragon
    0.71
    hood
    0.70
    sworth
    0.69
     Species
    0.69
    irez
    0.64
     Pau
    0.63
    Activations Density 0.008%

    No Known Activations