© 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. 15619
    Prev
    Next
    INDEX
    Explanations

    descriptions related to photography and camera equipment features

    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
    363
    +0.10
    0.5%
    382
    +0.10
    0.5%
    237
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    482
    +0.10
    0.01
    363
    +0.10
    0.01
    48
    +0.09
    0.01
    Negative Logits
    umbnails
    -0.71
     Bezos
    -0.69
    unal
    -0.62
     Elections
    -0.62
     Mason
    -0.61
     Mayer
    -0.60
     bribery
    -0.60
    aleb
    -0.60
    alis
    -0.60
     Keller
    -0.60
    POSITIVE LOGITS
     withd
    1.36
     WITH
    1.31
     with
    1.25
    with
    1.23
    WithNo
    1.13
    With
    1.08
     With
    1.01
    without
    0.99
     WITHOUT
    0.94
     without
    0.91
    Activations Density 0.084%

    No Known Activations