© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 2376
    Prev
    Next
    INDEX
    Explanations

    phrases related to confrontation and arguments, especially involving authority figures

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    626
    +0.11
    0.5%
    447
    +0.11
    0.5%
    373
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    628
    +0.11
    0.04
    25
    +0.11
    0.04
    254
    +0.11
    0.04
    Negative Logits
     Wikimedia
    -0.69
    Pg
    -0.67
    ibal
    -0.63
     quot
    -0.57
    bourg
    -0.56
    malink
    -0.55
    fred
    -0.55
    Thumbnail
    -0.54
    cision
    -0.54
    ITED
    -0.52
    POSITIVE LOGITS
     orally
    0.87
     beforehand
    0.78
     alike
    0.78
    DERR
    0.75
     how
    0.74
     goodbye
    0.74
    ij士
    0.71
     why
    0.70
    ounce
    0.69
     beware
    0.68
    Activations Density 0.224%

    No Known Activations