© 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. 5761
    Prev
    Next
    INDEX
    Explanations

    the word "that"

    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
    +0.14
    0.6%
    161
    +0.12
    0.6%
    274
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.14
    0.05
    161
    +0.12
    0.04
    655
    +0.12
    0.04
    Negative Logits
    WARD
    -0.66
    emis
    -0.64
    PRESS
    -0.62
    Desk
    -0.61
    Veter
    -0.60
    èĪ
    -0.60
    Luck
    -0.59
     Prepar
    -0.57
    Tax
    -0.57
    PP
    -0.57
    POSITIVE LOGITS
    cher
    0.97
    chers
    0.91
    ched
    0.82
     fateful
    0.80
     pesky
    0.78
    ching
    0.78
     damned
    0.72
     lasted
    0.71
     fundament
    0.67
     sort
    0.66
    Activations Density 0.144%

    No Known Activations