© 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 12288
    9. 8-RES_FS12288-JB
    11. 1718
    Prev
    Next
    INDEX
    Explanations

    phrases related to technology, products, social media, and popular culture

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    167
    +0.14
    0.7%
    608
    +0.13
    0.6%
    301
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.14
    0.06
    362
    +0.13
    0.06
    608
    +0.13
    0.06
    Negative Logits
    conom
    -0.81
    ppings
    -0.75
    cum
    -0.71
    bender
    -0.70
    puff
    -0.68
    arth
    -0.68
    ussen
    -0.67
     appears
    -0.67
     preceded
    -0.67
     belongs
    -0.66
    POSITIVE LOGITS
    selves
    1.17
     own
    1.11
     beloved
    0.90
     newest
    0.89
     motto
    0.88
     respective
    0.86
     ourselves
    0.82
     fearless
    0.80
     esteemed
    0.80
     collective
    0.80
    Activations Density 0.125%

    No Known Activations