© 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. 3828
    Prev
    Next
    INDEX
    Explanations

    news articles and related information regarding various events and updates

    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
    373
    +0.20
    0.9%
    69
    +0.11
    0.5%
    145
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.20
    -0.01
    23
    +0.11
    0.03
    145
    +0.10
    0.03
    Negative Logits
    represented
    -0.67
     subclass
    -0.61
    apter
    -0.61
    oteric
    -0.58
     earned
    -0.57
    alone
    -0.57
    Query
    -0.56
     bounded
    -0.56
     colored
    -0.55
    gener
    -0.55
    POSITIVE LOGITS
    htaking
    0.88
    :'
    0.81
    ÅŁ
    0.76
     Duterte
    0.75
    oÄŁan
    0.75
    ':
    0.71
     Thousands
    0.71
    ÄŁ
    0.68
    arnaev
    0.68
     SHARES
    0.66
    Activations Density 0.051%

    No Known Activations