© 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. 15446
    Prev
    Next
    INDEX
    Explanations

    phrases related to political appointments and nominations

    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
    483
    +0.11
    0.5%
    103
    +0.10
    0.5%
    755
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.11
    0.03
    23
    +0.10
    0.03
    352
    +0.09
    0.03
    Negative Logits
     seiz
    -0.71
    ãĥ¼ãĥĨ
    -0.66
     horizont
    -0.65
     srfAttach
    -0.62
     specificity
    -0.61
    pmwiki
    -0.60
    redients
    -0.59
     redes
    -0.58
     logger
    -0.58
     microphones
    -0.58
    POSITIVE LOGITS
    ĸļ
    1.15
    ı
    1.10
    ł
    1.10
    į
    1.09
    ¬
    1.08
    Ĵ
    1.07
    º
    1.06
    ¹
    1.04
    ĸ
    1.03
    ¡
    1.01
    Activations Density 0.040%

    No Known Activations