© 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. 15594
    Prev
    Next
    INDEX
    Explanations

    word pairs related to mathematics and scientific concepts

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.18
    0.9%
    678
    +0.13
    0.6%
    27
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.18
    0.06
    570
    +0.13
    0.02
    289
    +0.10
    0.02
    Negative Logits
    Office
    -0.70
     whistlebl
    -0.69
    EStream
    -0.68
    camp
    -0.68
     fundra
    -0.68
    Mobile
    -0.68
     vets
    -0.67
    Nin
    -0.67
    2018
    -0.67
    Camp
    -0.66
    POSITIVE LOGITS
     \(
    1.08
     atoms
    1.08
     equilibrium
    1.01
     finite
    0.99
     molecule
    0.96
     \(\
    0.95
     matrix
    0.95
     latt
    0.94
     ÎĶ
    0.93
     adjacent
    0.92
    Activations Density 0.476%

    No Known Activations