© 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. 16402
    Prev
    Next
    INDEX
    Explanations

    the word "to" followed by a high-ranking entity or action

    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
    393
    +0.12
    0.5%
    608
    +0.11
    0.5%
    143
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    104
    +0.12
    0.03
    608
    +0.11
    0.03
    364
    +0.10
    0.03
    Negative Logits
    VPN
    -0.70
     inferred
    -0.68
    Decl
    -0.64
     sake
    -0.63
    redict
    -0.62
    Secondly
    -0.62
    ially
    -0.61
    erity
    -0.61
    sych
    -0.61
    Got
    -0.60
    POSITIVE LOGITS
     haunt
    1.14
     shore
    0.90
     earth
    0.88
     fray
    0.86
     prominence
    0.86
     obscurity
    0.78
     Earth
    0.77
     fold
    0.77
     shores
    0.72
     Wonderland
    0.72
    Activations Density 0.059%

    No Known Activations