© 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. 1-RES-JB
    11. 21680
    Prev
    Next
    INDEX
    Explanations

    mentions of various organizations, such as schools and businesses, particularly those with the abbreviation "HS" or "TS" in their names

    oai_token-act-pair · gpt-3.5-turbo

    references to high schools and specific high school-related topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SourceFile
    -0.82
    netflix
    -0.81
    vertising
    -0.73
    perty
    -0.73
    rums
    -0.73
    rive
    -0.72
    ingen
    -0.72
    rontal
    -0.72
    amental
    -0.72
    rences
    -0.71
    POSITIVE LOGITS
     HS
    0.96
    ZI
    0.93
    TERN
    0.86
    peria
    0.77
    vir
    0.73
    ENS
    0.72
    TER
    0.71
    HT
    0.71
    uated
    0.68
     Blaz
    0.68
    Activations Density 0.023%

    No Known Activations