© 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. 14740
    Prev
    Next
    INDEX
    Explanations

    exact matches

    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
    362
    +0.13
    0.6%
    7
    +0.13
    0.6%
    158
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    158
    +0.13
    0.02
    347
    +0.13
    0.02
    7
    +0.11
    0.02
    Negative Logits
    udder
    -0.79
    Downloadha
    -0.78
    anners
    -0.76
    imaru
    -0.76
    ipolar
    -0.74
    ulton
    -0.73
     ashore
    -0.72
    ker
    -0.72
     Flavoring
    -0.72
    OPLE
    -0.71
    POSITIVE LOGITS
     wording
    1.04
    itude
    1.03
     opposite
    0.99
    ing
    0.98
    ions
    0.97
     same
    0.92
     embodiment
    0.89
     amount
    0.89
     number
    0.86
    ness
    0.84
    Activations Density 0.021%

    No Known Activations