© 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. 3790
    Prev
    Next
    INDEX
    Explanations

    code snippets or programming functions

    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
    392
    +0.12
    0.5%
    311
    +0.11
    0.5%
    152
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    392
    +0.12
    0.02
    356
    +0.11
    0.01
    311
    +0.11
    0.01
    Negative Logits
     Boll
    -0.77
     Thief
    -0.72
    assies
    -0.68
    jee
    -0.65
    akeru
    -0.65
     Incarn
    -0.61
     Madness
    -0.61
     Holly
    -0.60
    worldly
    -0.60
    ovych
    -0.59
    POSITIVE LOGITS
    initions
    1.41
    initely
    1.14
    icient
    1.00
    erence
    1.00
    endants
    0.98
    acement
    0.97
    inately
    0.95
    erred
    0.95
    inite
    0.94
    ensible
    0.94
    Activations Density 0.007%

    No Known Activations