© 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 3072
    9. 8-RES_FS3072-JB
    11. 778
    Prev
    Next
    INDEX
    Explanations

    mentions of copyrights and images in news articles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    373
    +0.11
    0.5%
    386
    +0.10
    0.4%
    739
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    559
    +0.11
    0.03
    121
    +0.10
    0.04
    739
    +0.10
    0.03
    Negative Logits
    ĪĴ
    -0.81
    OHN
    -0.72
     fortun
    -0.71
    Honest
    -0.68
    rency
    -0.67
    nesia
    -0.65
    apeshifter
    -0.64
    cember
    -0.64
     bowling
    -0.63
     lull
    -0.63
    POSITIVE LOGITS
     IMAGES
    0.87
     Image
    0.83
     Getty
    0.82
    UTERS
    0.77
     caption
    0.75
     AFP
    0.75
     Previous
    0.75
     Thousands
    0.75
    aceutical
    0.74
     Scenes
    0.72
    Activations Density 0.286%

    No Known Activations