© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3349
    Prev
    Next
    INDEX
    Explanations

    terms associated with political events and controversies

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    87
    +0.33
    1.7%
    756
    +0.12
    0.6%
    459
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.33
    -0.02
    119
    +0.12
    0.05
    8
    +0.11
    0.06
    Negative Logits
    icipated
    -0.67
    moil
    -0.66
    UGC
    -0.65
    etheless
    -0.64
    ÃįÃį
    -0.61
    ¶
    -0.60
    DCS
    -0.60
    TOR
    -0.59
    Following
    -0.58
    Amid
    -0.57
    POSITIVE LOGITS
     irony
    0.88
     kidding
    0.87
     analogy
    0.81
     fuckin
    0.81
     quotation
    0.77
     math
    0.76
     sir
    0.76
     joke
    0.75
     ital
    0.75
     guess
    0.73
    Activations Density 0.804%

    No Known Activations