© 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. 24416
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a nominated position or role

    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
    373
    +0.15
    0.7%
    665
    +0.14
    0.6%
    718
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.15
    0.04
    608
    +0.14
    0.03
    557
    +0.11
    0.03
    Negative Logits
    alore
    -0.80
    ratulations
    -0.69
    got
    -0.66
     reliant
    -0.66
    tested
    -0.66
    pointers
    -0.65
    browser
    -0.64
     surprises
    -0.63
     fireworks
    -0.63
    lets
    -0.63
    POSITIVE LOGITS
     oversee
    0.99
     testify
    0.84
     participate
    0.82
     bolster
    0.80
     serve
    0.80
     accompany
    0.79
     replace
    0.79
    姫
    0.79
     uphold
    0.78
     represent
    0.77
    Activations Density 0.106%

    No Known Activations