© 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 49152
    9. 8-RES_FS49152-JB
    11. 25950
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "to."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    608
    +0.14
    0.7%
    723
    +0.12
    0.6%
    123
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.14
    0.04
    104
    +0.12
    0.03
    665
    +0.12
    0.03
    Negative Logits
     hurled
    -0.72
    ŃĶ
    -0.72
    soever
    -0.71
    Ĥª
    -0.71
    hya
    -0.67
     mysteries
    -0.66
     glimps
    -0.66
    spir
    -0.65
     practition
    -0.65
    Words
    -0.63
    POSITIVE LOGITS
     accommodate
    0.88
     eliminate
    0.84
    oult
    0.84
     suppress
    0.80
     capitalize
    0.80
     consolidate
    0.77
     prioritize
    0.75
    ggles
    0.75
     Tycoon
    0.74
     exclude
    0.73
    Activations Density 0.060%

    No Known Activations