© 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. 20541
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to different topics or domains

    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
    +0.34
    1.6%
    480
    +0.20
    1.0%
    87
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.34
    0.03
    480
    +0.20
    0.01
    101
    +0.12
    0.04
    Negative Logits
    luster
    -0.74
    raints
    -0.74
    nesday
    -0.69
    terday
    -0.67
    onto
    -0.62
    ecided
    -0.61
    aeda
    -0.60
     outwe
    -0.59
    coni
    -0.59
    ainment
    -0.58
    POSITIVE LOGITS
     âĢº
    0.69
     ·
    0.67
     COURT
    0.65
     Profile
    0.61
    é¾į
    0.59
     Photograph
    0.59
     largeDownload
    0.58
     Files
    0.57
     translator
    0.56
     Kills
    0.55
    Activations Density 0.167%

    No Known Activations