© 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. 3201
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal and professional backgrounds including careers and accomplishments

    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
    727
    +0.12
    0.5%
    373
    +0.11
    0.5%
    112
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.12
    0.07
    727
    +0.11
    0.04
    545
    +0.09
    0.04
    Negative Logits
     dstg
    -0.82
    wayne
    -0.79
    abouts
    -0.68
    taboola
    -0.67
    dayName
    -0.66
    pmwiki
    -0.62
    nm
    -0.62
    inement
    -0.61
    aminer
    -0.61
    peat
    -0.61
    POSITIVE LOGITS
     prominence
    0.94
     notoriety
    0.87
     fame
    0.81
     recently
    0.70
     decom
    0.68
     Challenger
    0.68
     shortly
    0.64
     independence
    0.63
     earliest
    0.60
     Flavoring
    0.59
    Activations Density 0.457%

    No Known Activations