© 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. 24445
    Prev
    Next
    INDEX
    Explanations

    numerical data, specifically percentages and statistics

    oai_token-act-pair · gpt-3.5-turbo

    numerical percentages indicating significant changes or results

    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
     Machines
    -0.77
     devils
    -0.73
     machines
    -0.72
     Halls
    -0.72
     IDs
    -0.70
     memories
    -0.69
     wills
    -0.67
    ATURES
    -0.67
     establishments
    -0.67
     avenues
    -0.66
    POSITIVE LOGITS
    %-
    1.12
    %
    1.03
     percent
    1.02
    percent
    0.96
     percentage
    0.93
     percentile
    0.93
     fold
    0.89
     rating
    0.89
    fold
    0.88
     multiplier
    0.88
    Activations Density 0.111%

    No Known Activations