© 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. 24099
    Prev
    Next
    INDEX
    Explanations

    numerical quantities and punctuation in various contexts

    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
    373
    +0.12
    0.5%
    447
    +0.10
    0.5%
    579
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    739
    +0.12
    0.03
    164
    +0.10
    0.04
    428
    +0.10
    0.03
    Negative Logits
    kefeller
    -0.85
     flo
    -0.74
    enhagen
    -0.70
    emort
    -0.68
    NFL
    -0.66
     conclud
    -0.65
    undy
    -0.64
    ipeg
    -0.63
    :(
    -0.62
     payoff
    -0.62
    POSITIVE LOGITS
     Galile
    0.93
     Saga
    0.90
     Municip
    0.88
    Äĩ
    0.85
     Converted
    0.85
     Catalonia
    0.83
    Ø©
    0.82
     Catalan
    0.81
     Goddess
    0.78
     Uzbek
    0.77
    Activations Density 0.499%

    No Known Activations