© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 7922
    Prev
    Next
    INDEX
    Explanations

    different documents or articles

    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
    373
    +0.10
    0.5%
    362
    +0.10
    0.5%
    716
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.04
    608
    +0.10
    0.04
    716
    +0.10
    0.03
    Negative Logits
    ogie
    -0.77
    ican
    -0.70
    Ore
    -0.69
    mares
    -0.68
     TAMADRA
    -0.67
    ilee
    -0.66
    buster
    -0.66
    finals
    -0.63
    erto
    -0.62
    Kings
    -0.60
    POSITIVE LOGITS
     standpoint
    1.68
     vantage
    1.47
     perspective
    1.45
     outset
    1.45
     perspectives
    1.25
     viewpoint
    1.23
     angles
    1.05
     inception
    0.95
     afar
    0.94
     angle
    0.94
    Activations Density 0.089%

    No Known Activations