© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 3072
    9. 8-RES_FS3072-JB
    11. 38
    Prev
    Next
    INDEX
    Explanations

    proper nouns, especially names of individuals

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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%
    562
    +0.09
    0.4%
    55
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    598
    +0.18
    0.06
    119
    +0.09
    0.08
    582
    +0.08
    0.06
    Negative Logits
    quartered
    -0.98
    itect
    -0.69
    rongh
    -0.68
    leading
    -0.67
    DN
    -0.65
    rendered
    -0.65
    hib
    -0.64
     Presence
    -0.64
    tex
    -0.63
    Required
    -0.63
    POSITIVE LOGITS
     begs
    0.87
     kicker
    0.85
     darn
    0.76
     prest
    0.75
     damned
    0.73
     damn
    0.71
     sucks
    0.70
     Cohn
    0.70
     luck
    0.69
     forgot
    0.68
    Activations Density 2.208%

    No Known Activations