© 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. 2990
    Prev
    Next
    INDEX
    Explanations

    phrases or statements indicating what should be done or required in a given situation

    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
    481
    +0.21
    1.0%
    480
    +0.20
    1.0%
    447
    +0.16
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    243
    +0.21
    0.02
    273
    +0.20
    0.02
    560
    +0.16
    0.01
    Negative Logits
     Kurd
    -0.70
     Niet
    -0.69
     Palestin
    -0.67
     Seym
    -0.66
     Azerb
    -0.64
    âĨij
    -0.63
     Stain
    -0.62
    edIn
    -0.62
     Examiner
    -0.60
     prest
    -0.60
    POSITIVE LOGITS
    heading
    0.83
    yond
    0.81
    ijing
    0.81
    hemoth
    0.81
    gins
    0.80
    jew
    0.78
    league
    0.75
    Í
    0.75
    ardless
    0.74
    eps
    0.72
    Activations Density 0.028%

    No Known Activations