© 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. 3855
    Prev
    Next
    INDEX
    Explanations

    phrases related to competition and support

    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
    226
    +0.13
    0.6%
    143
    +0.11
    0.5%
    536
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    212
    +0.13
    0.03
    143
    +0.11
    0.03
    326
    +0.09
    0.05
    Negative Logits
    option
    -0.69
    dx
    -0.66
    Corp
    -0.65
    ascript
    -0.61
    berus
    -0.59
    enda
    -0.59
    ashington
    -0.57
    osuke
    -0.57
    GI
    -0.57
    odi
    -0.56
    POSITIVE LOGITS
     them
    0.94
     their
    0.87
     THEM
    0.87
     THEIR
    0.84
     themselves
    0.83
     us
    0.76
     our
    0.75
    hyde
    0.74
    lessly
    0.74
    edIn
    0.72
    Activations Density 0.368%

    No Known Activations