© 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. 5158
    Prev
    Next
    INDEX
    Explanations

    phrases related to social events and gatherings

    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
    714
    +0.12
    0.6%
    408
    +0.11
    0.5%
    393
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    484
    +0.12
    0.06
    714
    +0.11
    0.04
    0
    +0.11
    0.05
    Negative Logits
    Favorite
    -0.81
    grand
    -0.79
    anu
    -0.78
    urden
    -0.77
    wic
    -0.77
    conduct
    -0.75
     Own
    -0.74
    raid
    -0.73
    uh
    -0.72
    ivot
    -0.71
    POSITIVE LOGITS
     consequently
    1.01
     therefore
    0.98
     hence
    0.96
     replaced
    0.96
     thus
    0.90
     nobody
    0.88
     there
    0.83
     pandemonium
    0.83
     everyone
    0.80
     everybody
    0.80
    Activations Density 0.245%

    No Known Activations