© 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. 8-RES-JB
    11. 10605
    Prev
    Next
    INDEX
    Explanations

    locations, especially buildings like houses, hospitals, and stores

    oai_token-act-pair · gpt-3.5-turbo

    places and institutions that are associated with public services and community activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    %%%%
    -0.64
    Trend
    -0.63
    ================================================================
    -0.60
     bets
    -0.57
    icated
    -0.57
    layer
    -0.55
    %).
    -0.55
    athered
    -0.55
    athi
    -0.55
     Fact
    -0.55
    POSITIVE LOGITS
     expecting
    0.87
     undet
    0.83
     where
    0.75
    itent
    0.72
     whence
    0.71
     fray
    0.71
     doorstep
    0.71
     Headquarters
    0.71
     headquarters
    0.69
     tremend
    0.65
    Activations Density 0.523%

    No Known Activations