© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 15853
    Prev
    Next
    INDEX
    Explanations

    mentions of legal regulations and laws

    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
    575
    +0.13
    0.6%
    481
    +0.12
    0.6%
    447
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.13
    0.03
    524
    +0.12
    0.02
    258
    +0.10
    0.02
    Negative Logits
    yss
    -0.76
    ience
    -0.70
    asus
    -0.67
    ortun
    -0.64
    stocks
    -0.64
    Tur
    -0.64
     Mous
    -0.63
    joy
    -0.62
    otal
    -0.62
    fortune
    -0.61
    POSITIVE LOGITS
     enforcement
    1.42
     prohibiting
    1.32
     governing
    1.31
     regulating
    1.30
     enacted
    1.19
    enforcement
    1.18
     enforcing
    1.12
     Enforcement
    1.11
    abiding
    1.11
     restricting
    1.11
    Activations Density 0.029%

    No Known Activations