© 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. 1742
    Prev
    Next
    INDEX
    Explanations

    GitHub and LinkedIn references in the text

    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
    508
    +0.12
    0.6%
    294
    +0.11
    0.5%
    447
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.12
    0.05
    508
    +0.11
    0.05
    294
    +0.11
    0.04
    Negative Logits
    ibus
    -0.79
    este
    -0.68
    ften
    -0.66
    ~~~~~~~~~~~~~~~~
    -0.64
    bery
    -0.62
    arching
    -0.61
    pherd
    -0.61
    alam
    -0.61
    hani
    -0.60
     Reagan
    -0.59
    POSITIVE LOGITS
     username
    0.91
    ®
    0.87
     Username
    0.85
    Connection
    0.85
    Maps
    0.84
     Commons
    0.83
     Labs
    0.83
     Analytics
    0.82
     Users
    0.81
     user
    0.81
    Activations Density 0.156%

    No Known Activations