© 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. 7611
    Prev
    Next
    INDEX
    Explanations

    themes related to personal endeavors, achievements, and activities

    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
    447
    +0.22
    1.0%
    481
    +0.17
    0.8%
    353
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    353
    +0.22
    0.04
    761
    +0.17
    0.04
    70
    +0.12
    0.04
    Negative Logits
     assurances
    -0.73
    ppings
    -0.68
     Orwell
    -0.66
    SPONSORED
    -0.65
     Nielsen
    -0.64
    imaru
    -0.62
    xit
    -0.62
     pretext
    -0.62
    Detailed
    -0.58
    eus
    -0.58
    POSITIVE LOGITS
     professionally
    1.08
     solo
    0.86
     blogging
    0.83
     freelance
    0.81
     arts
    0.80
     gigs
    0.80
     comics
    0.79
     podcasts
    0.77
     hobby
    0.77
     photography
    0.77
    Activations Density 0.280%

    No Known Activations