© 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. 2995
    Prev
    Next
    INDEX
    Explanations

    government officials' names

    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
    430
    +0.09
    0.4%
    521
    +0.09
    0.4%
    98
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.09
    0.05
    232
    +0.09
    0.04
    466
    +0.09
    0.04
    Negative Logits
    pmwiki
    -0.75
    ãĥ¼ãĥĨ
    -0.74
    DragonMagazine
    -0.74
     notebooks
    -0.71
    zsche
    -0.69
     interpol
    -0.68
    CVE
    -0.68
    Alien
    -0.66
    ãĤ»
    -0.66
    swick
    -0.65
    POSITIVE LOGITS
     Mans
    0.76
     appoint
    0.76
    itol
    0.76
     Legislature
    0.73
     appointed
    0.71
     Hancock
    0.71
    llor
    0.69
    bernatorial
    0.68
    HUD
    0.68
     Governor
    0.63
    Activations Density 0.116%

    No Known Activations