© 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. 1695
    Prev
    Next
    INDEX
    Explanations

    phrases related to causes or origins

    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
    759
    +0.13
    0.6%
    252
    +0.12
    0.6%
    331
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    252
    +0.13
    0.06
    759
    +0.12
    0.05
    569
    +0.09
    0.05
    Negative Logits
    ahs
    -0.77
    mouth
    -0.72
    hold
    -0.65
     dumps
    -0.65
     tumble
    -0.59
     requested
    -0.59
    ugh
    -0.59
     Ich
    -0.59
    ãĥīãĥ©
    -0.58
    dogs
    -0.58
    POSITIVE LOGITS
     partly
    1.04
     solely
    0.99
     chiefly
    0.93
     principally
    0.90
     squarely
    0.88
     primarily
    0.82
     purely
    0.79
    actionDate
    0.77
     largely
    0.77
     mainly
    0.76
    Activations Density 0.372%

    No Known Activations