© 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 24576
    9. 8-RES_FS24576-JB
    11. 7061
    Prev
    Next
    INDEX
    Explanations

    mentions of data privacy issues

    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
    373
    +0.17
    0.9%
    60
    +0.08
    0.4%
    426
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.17
    -0.01
    659
    +0.08
    0.03
    639
    +0.08
    0.03
    Negative Logits
    ovich
    -0.57
    reddits
    -0.54
     SHARES
    -0.53
    é¾
    -0.51
    owsky
    -0.51
    é¾įå
    -0.50
     Donation
    -0.49
     IMAGES
    -0.49
     Volunteers
    -0.48
    aukee
    -0.48
    POSITIVE LOGITS
    uce
    0.68
    via
    0.64
    zzo
    0.64
    gg
    0.62
    versible
    0.61
    agues
    0.61
    aded
    0.61
    ading
    0.60
    bra
    0.60
    usted
    0.60
    Activations Density 0.093%

    No Known Activations