© 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. 23029
    Prev
    Next
    INDEX
    Explanations

    instances of articles and pronouns

    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
    36
    +0.26
    1.2%
    250
    +0.12
    0.6%
    393
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.26
    0.12
    250
    +0.12
    0.06
    263
    +0.12
    0.05
    Negative Logits
     indign
    -0.77
     appell
    -0.71
    anism
    -0.69
    livious
    -0.67
    establishment
    -0.67
    forth
    -0.66
     impat
    -0.65
     airs
    -0.65
     indoctr
    -0.65
    enance
    -0.64
    POSITIVE LOGITS
     bunch
    0.99
     couple
    0.98
    uras
    0.92
     few
    0.91
     lot
    0.89
     guest
    0.88
    ria
    0.86
     copy
    0.86
     friend
    0.83
    uctions
    0.82
    Activations Density 0.228%

    No Known Activations