© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 9323
    Prev
    Next
    INDEX
    Explanations

    assertions of worth, skill, or love between individuals

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    747
    +0.12
    0.5%
    39
    +0.09
    0.4%
    707
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.06
    567
    +0.09
    0.04
    92
    +0.09
    0.04
    Negative Logits
     Shutterstock
    -0.73
     Springer
    -0.72
    backs
    -0.67
    letal
    -0.65
     NCT
    -0.62
     discussing
    -0.62
    ussen
    -0.61
    utan
    -0.61
    pport
    -0.61
    strate
    -0.61
    POSITIVE LOGITS
     validity
    1.33
     authenticity
    1.31
     existence
    1.30
     innocence
    1.26
    worthiness
    1.23
     legitimacy
    1.22
     correctness
    1.19
     superiority
    1.10
    iability
    1.04
     usefulness
    1.03
    Activations Density 0.400%

    No Known Activations