© 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. 22809
    Prev
    Next
    INDEX
    Explanations

    words related to agency or actions performed by individuals

    oai_token-act-pair · gpt-4o-miniTriggered 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
    362
    +0.11
    0.5%
    658
    +0.11
    0.5%
    677
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.04
    22
    +0.11
    0.03
    616
    +0.11
    0.03
    Negative Logits
     enjoys
    -0.81
     accepts
    -0.80
     reacts
    -0.78
     assumes
    -0.73
     acknowledges
    -0.69
     observes
    -0.68
     understands
    -0.67
     inspires
    -0.67
     participates
    -0.66
     foregoing
    -0.65
    POSITIVE LOGITS
    lined
    0.89
    lain
    0.74
    oby
    0.73
    activated
    0.71
    seless
    0.70
    vol
    0.69
    tained
    0.68
    oteric
    0.68
    drawn
    0.67
    inently
    0.66
    Activations Density 0.185%

    No Known Activations