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

    Neuronpedia

    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. 1628
    Prev
    Next
    INDEX
    Explanations

    terms related to events or conferences that are named "Summit"

    oai_token-act-pair · gpt-3.5-turbo

    references to various summits or conferences

    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
     gir
    -0.77
     fishes
    -0.77
     rob
    -0.72
     pract
    -0.71
     pir
    -0.71
     retali
    -0.71
    REL
    -0.70
     hitherto
    -0.70
     oblig
    -0.68
     fer
    -0.68
    POSITIVE LOGITS
     Summit
    3.80
    Sum
    1.18
     Expo
    1.17
     Surge
    1.10
     Conference
    1.08
     Semin
    1.00
     Alliance
    0.96
     Junction
    0.95
     Accord
    0.94
     Harbor
    0.92
    Activations Density 0.012%

    No Known Activations