© 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 EvalsExportsinterp-engineNEWAPI 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. 20900
    Prev
    Next
    INDEX
    Explanations

    locations or settings where specific events occur with a focus on action-related words like "was," "spotted," and "saw."

    oai_token-act-pair · gpt-3.5-turbo

    situations involving a specific event occurring at a particular time

    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
    Score
    -0.73
    2020
    -0.70
    peat
    -0.70
    ouble
    -0.69
    ETF
    -0.68
    gem
    -0.67
    ãĥĺ
    -0.67
    medi
    -0.66
    ãĥ³ãĤ¸
    -0.65
    ĸļ
    -0.64
    POSITIVE LOGITS
     suddenly
    1.15
     confronted
    1.00
     gunmen
    0.96
     he
    0.90
     she
    0.89
     gunfire
    0.88
     someone
    0.87
     approached
    0.87
     something
    0.86
     gunshots
    0.86
    Activations Density 0.137%

    No Known Activations