© 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 49152
    9. 8-RES_FS49152-JB
    11. 45791
    Prev
    Next
    INDEX
    Explanations

    names of prominent writers and intellectuals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.10
    0.5%
    41
    +0.09
    0.4%
    441
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.04
    525
    +0.09
    0.03
    556
    +0.09
    0.02
    Negative Logits
    esville
    -0.73
    omore
    -0.69
    sterdam
    -0.68
    ataka
    -0.67
    imity
    -0.66
    natureconservancy
    -0.66
    alach
    -0.66
    aryn
    -0.64
    RGB
    -0.63
    IFA
    -0.62
    POSITIVE LOGITS
    ãĥ¼ãĥĨ
    0.71
     III
    0.67
     Cable
    0.66
     Poe
    0.64
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.64
     Lenin
    0.63
     Roberts
    0.63
     Curtis
    0.62
     Sullivan
    0.61
     Pers
    0.61
    Activations Density 0.057%

    No Known Activations