© 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 EvalsExportsAPI 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. 7741
    Prev
    Next
    INDEX
    Explanations

    references to specific terms related to video games and entertainment media

    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
    447
    +0.21
    1.0%
    481
    +0.12
    0.6%
    314
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.21
    0.07
    314
    +0.12
    0.01
    725
    +0.11
    0.05
    Negative Logits
     Aber
    -0.73
     arrang
    -0.67
     Offer
    -0.66
     [&
    -0.63
     Delay
    -0.62
     Merrill
    -0.62
     Goldstein
    -0.62
     guiName
    -0.61
     Shirley
    -0.60
     timet
    -0.60
    POSITIVE LOGITS
    ikuman
    0.72
    uterte
    0.72
    aji
    0.71
    pedia
    0.71
    otech
    0.67
    itar
    0.67
    obos
    0.67
    ageddon
    0.66
    zee
    0.66
    razil
    0.66
    Activations Density 0.352%

    No Known Activations