© 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. 11-RES-JB
    11. 20620
    Prev
    Next
    INDEX
    Explanations

    terms related to numbers, quantities, and various items or entities

    oai_token-act-pair · gpt-3.5-turbo

    references to diverse groups and sources in a context involving data collection or funding

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enario
    -0.65
     Sorceress
    -0.64
     takeoff
    -0.64
    kamp
    -0.64
     Archdemon
    -0.64
    ilda
    -0.63
    eanor
    -0.62
    ander
    -0.62
    anship
    -0.62
    acerb
    -0.62
    POSITIVE LOGITS
     sources
    1.55
     organizations
    1.48
     entities
    1.43
     outlets
    1.42
     agencies
    1.41
     vendors
    1.40
     suppliers
    1.39
     organisations
    1.38
     parties
    1.38
     providers
    1.33
    Activations Density 0.355%

    No Known Activations