© 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. 10045
    Prev
    Next
    INDEX
    Explanations

    company names starting with "Deloitte" and mentions specific details about their reports or services

    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
    373
    +0.13
    0.6%
    712
    +0.11
    0.5%
    314
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.13
    -0.01
    659
    +0.11
    0.03
    232
    +0.09
    0.02
    Negative Logits
    belt
    -0.61
    atsu
    -0.60
    prem
    -0.60
    Ô
    -0.56
    Austral
    -0.56
     bout
    -0.55
     round
    -0.54
     bour
    -0.53
     Mubarak
    -0.53
     BASE
    -0.53
    POSITIVE LOGITS
    ciating
    1.11
    icio
    0.87
    idated
    0.85
    ij士
    0.80
    ãĤ¼ãĤ¦ãĤ¹
    0.74
    escent
    0.72
    ierre
    0.71
    oppers
    0.70
    owship
    0.70
    itte
    0.66
    Activations Density 0.069%

    No Known Activations