© 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. 7551
    Prev
    Next
    INDEX
    Explanations

    studies and research findings related to various societal and health 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
    447
    +0.12
    0.5%
    679
    +0.12
    0.5%
    375
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    375
    +0.12
    0.03
    417
    +0.12
    0.03
    747
    +0.11
    0.03
    Negative Logits
    arta
    -0.73
    EStreamFrame
    -0.67
    awar
    -0.66
    perty
    -0.65
    brance
    -0.63
    quer
    -0.63
    cius
    -0.61
    clock
    -0.60
    assetsadobe
    -0.58
    press
    -0.57
    POSITIVE LOGITS
     traces
    0.89
     evidence
    0.80
     conclusive
    0.77
     correlations
    0.77
     significant
    0.75
     discrepancies
    0.74
    ered
    0.74
     widespread
    0.73
     fault
    0.73
    INGS
    0.70
    Activations Density 0.054%

    No Known Activations