© 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. 4-RES-JB
    11. 1240
    Prev
    Next
    INDEX
    Explanations

    mentions of specific physical locations or sites

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific geographical or archaeological sites

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

    No Comments

    Negative Logits
    ornia
    -0.86
    atures
    -0.76
    olyn
    -0.73
    cia
    -0.70
     Leilan
    -0.68
    icipated
    -0.68
     actionGroup
    -0.67
    jri
    -0.66
    sen
    -0.65
    BuyableInstoreAndOnline
    -0.64
    POSITIVE LOGITS
     sites
    0.86
     site
    0.86
    builders
    0.82
     Site
    0.80
    builder
    0.75
     redevelopment
    0.74
    ologies
    0.72
     tenant
    0.71
     location
    0.70
    bench
    0.70
    Activations Density 0.024%

    No Known Activations