© 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. 10640
    Prev
    Next
    INDEX
    Explanations

    places and organizations

    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
    546
    +0.14
    0.6%
    140
    +0.09
    0.4%
    616
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    546
    +0.14
    0.10
    297
    +0.09
    0.07
    140
    +0.09
    0.07
    Negative Logits
    itus
    -0.69
    raph
    -0.67
    orem
    -0.65
     impedance
    -0.64
    ionic
    -0.64
    iple
    -0.63
    ptin
    -0.63
    orph
    -0.63
    prints
    -0.62
    inct
    -0.62
    POSITIVE LOGITS
    where
    1.22
     where
    1.19
    England
    1.01
     Chile
    0.87
     Netherlands
    0.86
     Belgium
    0.86
     Territories
    0.85
     birthplace
    0.85
     England
    0.84
     USA
    0.84
    Activations Density 0.470%

    No Known Activations