© 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 24576
    9. 8-RES_FS24576-JB
    11. 22251
    Prev
    Next
    INDEX
    Explanations

    present tense forms of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered 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
    418
    +0.10
    0.5%
    102
    +0.10
    0.5%
    7
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.10
    0.09
    7
    +0.10
    0.06
    566
    +0.10
    0.06
    Negative Logits
    \\\\\\\\
    -0.74
    iHUD
    -0.72
    sburg
    -0.65
     viz
    -0.65
    west
    -0.65
    00007
    -0.64
    2020
    -0.64
     conviction
    -0.63
     Tours
    -0.62
    LOG
    -0.62
    POSITIVE LOGITS
     prolifer
    1.20
     everywhere
    1.08
     notoriously
    1.03
     plentiful
    0.99
     typically
    0.98
    senal
    0.95
     rarely
    0.95
     commonly
    0.93
     generally
    0.93
     inherently
    0.93
    Activations Density 0.292%

    No Known Activations