© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 5604
    Prev
    Next
    INDEX
    Explanations

    phrases related to governance, democracy, and responsibilities as citizens

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    393
    +0.12
    0.6%
    756
    +0.12
    0.5%
    562
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.12
    0.05
    139
    +0.12
    0.03
    449
    +0.10
    0.03
    Negative Logits
    mma
    -0.60
     Zap
    -0.59
     Downing
    -0.59
     Drunk
    -0.59
     Dawson
    -0.56
    Jar
    -0.56
    Dri
    -0.56
     Thanksgiving
    -0.56
     Belfast
    -0.56
     Varg
    -0.56
    POSITIVE LOGITS
    iety
    0.91
     aspire
    0.82
     ought
    0.78
    nery
    0.75
     hereafter
    0.75
    abouts
    0.73
     inspires
    0.71
    Else
    0.71
    selves
    0.70
    else
    0.69
    Activations Density 0.226%

    No Known Activations