© 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. 21270
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing a specific point or aspect, often starting with "the one."

    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
    408
    +0.10
    0.4%
    609
    +0.10
    0.4%
    688
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.10
    0.03
    436
    +0.10
    0.02
    171
    +0.10
    0.03
    Negative Logits
    mson
    -0.74
    iolet
    -0.71
    inders
    -0.71
    ocks
    -0.70
    avascript
    -0.67
    ories
    -0.66
    azard
    -0.65
    ooks
    -0.65
    illus
    -0.64
    otonin
    -0.63
    POSITIVE LOGITS
     thing
    0.94
     hundred
    0.92
     exception
    0.91
     Hundred
    0.85
     sided
    0.79
     eyed
    0.76
     guy
    0.74
     thousand
    0.73
     drawback
    0.71
     caveat
    0.70
    Activations Density 0.048%

    No Known Activations