© 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. 0-RES-JB
    11. 12732
    Prev
    Next
    INDEX
    Explanations

    references to specific elements: uranium, Chang's defection, and the timeline leading to the capability of developing a nuclear bomb

    oai_token-act-pair · gpt-3.5-turbo

    references to specific animals, particularly insects or arachnids

    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
     chanted
    -0.66
    âĶĢâĶĢ
    -0.65
     Guth
    -0.63
     Satoshi
    -0.63
    displayText
    -0.60
    ository
    -0.60
    CRIPTION
    -0.59
     Founding
    -0.59
     Colo
    -0.59
     Kyoto
    -0.59
    POSITIVE LOGITS
    brates
    0.81
    oggles
    0.80
    rate
    0.70
    zl
    0.70
    helm
    0.69
    merce
    0.67
     perm
    0.66
    flo
    0.66
    brate
    0.65
    iasis
    0.65
    Activations Density 0.000%

    No Known Activations