© 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. 3880
    Prev
    Next
    INDEX
    Explanations

    the word "past."

    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
    +0.15
    0.7%
    76
    +0.11
    0.5%
    284
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    76
    +0.15
    0.03
    284
    +0.11
    0.03
    199
    +0.09
    0.03
    Negative Logits
     Franch
    -0.76
    lees
    -0.75
    hyde
    -0.71
    },{"
    -0.70
    NEY
    -0.67
    racuse
    -0.66
    csv
    -0.65
    ODY
    -0.64
    iform
    -0.64
    iquid
    -0.61
    POSITIVE LOGITS
    ebin
    1.43
     decade
    1.34
     tense
    1.02
     few
    1.01
     millennium
    1.00
     century
    1.00
     fortnight
    0.99
     couple
    0.99
     week
    0.99
     month
    0.93
    Activations Density 0.021%

    No Known Activations