© 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. 16498
    Prev
    Next
    INDEX
    Explanations

    phrases related to political figures and their actions

    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
    87
    +0.31
    1.6%
    435
    +0.09
    0.5%
    503
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.31
    -0.03
    665
    +0.09
    0.08
    570
    +0.09
    0.02
    Negative Logits
     Culture
    -0.56
    ogun
    -0.56
    inus
    -0.54
     Freddie
    -0.52
    cies
    -0.51
    @#&
    -0.51
    displayText
    -0.51
    ĸļ
    -0.50
    sels
    -0.49
    olulu
    -0.48
    POSITIVE LOGITS
     someday
    1.35
     tomorrow
    0.84
     anytime
    0.84
     depending
    0.80
     sooner
    0.80
     future
    0.76
     if
    0.75
     undet
    0.75
     somew
    0.73
    depending
    0.70
    Activations Density 1.113%

    No Known Activations