© 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. 3319
    Prev
    Next
    INDEX
    Explanations

    phrases related to objects or concepts located above or in the air

    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
    570
    +0.11
    0.5%
    632
    +0.11
    0.5%
    481
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.03
    99
    +0.11
    0.02
    693
    +0.10
    0.02
    Negative Logits
    abul
    -0.91
    lda
    -0.90
    erion
    -0.88
    abase
    -0.84
    bard
    -0.84
    ku
    -0.83
    aido
    -0.83
    uther
    -0.82
    oshenko
    -0.82
    io
    -0.81
    POSITIVE LOGITS
     aerial
    0.99
     crane
    0.84
     viewing
    0.83
     observation
    0.81
     telephone
    0.79
     view
    0.76
     projector
    0.76
     spraying
    0.76
     photographs
    0.76
     examination
    0.75
    Activations Density 0.057%

    No Known Activations