© 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. 10473
    Prev
    Next
    INDEX
    Explanations

    commands or instructions in a technical context

    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
    640
    +0.17
    0.8%
    312
    +0.10
    0.5%
    434
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.17
    0.05
    312
    +0.10
    0.03
    160
    +0.09
    0.03
    Negative Logits
     Mechdragon
    -0.75
    hement
    -0.73
     Zig
    -0.71
    hao
    -0.67
     Scion
    -0.66
     Cycle
    -0.64
     Kinnikuman
    -0.63
     Ripple
    -0.61
    hing
    -0.60
     Kirin
    -0.60
    POSITIVE LOGITS
    sonian
    0.82
    ³³³³³³³³
    0.78
    nature
    0.76
    mob
    0.75
    equal
    0.75
    ³³³³³³³³³³³³³³³³
    0.74
     nonex
    0.70
    similar
    0.69
    fee
    0.69
    ever
    0.69
    Activations Density 0.143%

    No Known Activations