© 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. 9-RES-JB
    11. 5423
    Prev
    Next
    INDEX
    Explanations

    dates and locations mentioned in news articles

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Berlin" followed by numerical values or formats typically associated with news headlines

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    halla
    -0.80
    haus
    -0.78
     rele
    -0.74
     ourselves
    -0.73
    ãĤ±
    -0.72
     carpet
    -0.66
     oneself
    -0.65
    sie
    -0.65
    shi
    -0.65
    bage
    -0.64
    POSITIVE LOGITS
     Thousands
    1.17
     Dozens
    1.07
     Hundreds
    1.03
     Nearly
    0.99
     Protesters
    0.93
     Several
    0.93
     Authorities
    0.92
     Prosecutors
    0.92
     Researchers
    0.92
     Despite
    0.91
    Activations Density 0.025%

    No Known Activations