© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3679
    Prev
    Next
    INDEX
    Explanations

    content related to software development and programming

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.16
    0.7%
    689
    +0.10
    0.5%
    419
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.16
    0.04
    689
    +0.10
    0.04
    540
    +0.10
    0.04
    Negative Logits
     ultraviolet
    -0.65
    onics
    -0.64
     ILCS
    -0.62
    $$$$
    -0.59
    stairs
    -0.58
    0010
    -0.56
     everyday
    -0.56
     Magicka
    -0.56
    liga
    -0.54
     scissors
    -0.53
    POSITIVE LOGITS
     installments
    0.85
     iterations
    0.81
     administrations
    0.81
    iations
    0.74
     eras
    0.73
    ceptions
    0.70
    frames
    0.68
     generations
    0.68
     unsuccessfully
    0.67
    bernatorial
    0.67
    Activations Density 0.243%

    No Known Activations