© 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. 5716
    Prev
    Next
    INDEX
    Explanations

    mentioning issues, names, players or tools

    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
    +0.17
    0.8%
    679
    +0.13
    0.6%
    361
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.17
    0.05
    455
    +0.13
    0.04
    276
    +0.11
    0.04
    Negative Logits
    ople
    -0.69
     ItemLevel
    -0.66
    [_
    -0.64
     "<
    -0.62
     Tsukuyomi
    -0.61
    ructose
    -0.60
    maximum
    -0.60
    Interstitial
    -0.58
    nai
    -0.57
    renheit
    -0.57
    POSITIVE LOGITS
     factor
    0.88
     consideration
    0.85
     besides
    0.82
    worldly
    0.81
     noteworthy
    0.79
     aspect
    0.77
    ilyn
    0.76
    factor
    0.74
     relates
    0.74
     pecul
    0.71
    Activations Density 0.208%

    No Known Activations