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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 22309
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly parentheses and periods

    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
    481
    +0.14
    0.6%
    241
    +0.12
    0.5%
    302
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    114
    +0.14
    0.03
    302
    +0.12
    0.04
    241
    +0.11
    0.03
    Negative Logits
    ãĥ¥
    -0.71
    ãĥ´ãĤ¡
    -0.65
    UCT
    -0.64
    ĸļ
    -0.64
     eg
    -0.60
     fibre
    -0.60
    ¥µ
    -0.58
    TOR
    -0.58
    Ĥª
    -0.58
     guiActiveUnfocused
    -0.57
    POSITIVE LOGITS
     vou
    0.70
     Jr
    0.67
     insists
    0.67
     resigned
    0.66
     wrote
    0.65
    hatt
    0.65
    gew
    0.64
     congratulated
    0.64
     famously
    0.64
     believes
    0.64
    Activations Density 0.088%

    No Known Activations