© 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. 5807
    Prev
    Next
    INDEX
    Explanations

    names of people, especially repeated names like "Norris" and "Trout"

    oai_token-act-pair · gpt-3.5-turbo

    prominent names of individuals, specifically in a crime-related context

    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
    endment
    -0.74
    olitan
    -0.71
    ilities
    -0.68
    mails
    -0.67
    iser
    -0.66
     Cycl
    -0.62
    bath
    -0.62
    fare
    -0.61
    front
    -0.60
    runner
    -0.60
    POSITIVE LOGITS
    ational
    0.85
    ilater
    0.83
    ormal
    0.81
    ensible
    0.81
     Gord
    0.75
     Institution
    0.74
    oku
    0.72
    orp
    0.71
    agar
    0.70
    rir
    0.70
    Activations Density 0.021%

    No Known Activations