© 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. 13962
    Prev
    Next
    INDEX
    Explanations

    words related to taking action or carrying out tasks

    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
    480
    +0.20
    1.0%
    447
    +0.17
    0.8%
    464
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    464
    +0.20
    0.04
    257
    +0.17
    0.03
    722
    +0.14
    0.03
    Negative Logits
     vulner
    -0.70
     Niet
    -0.67
     Seym
    -0.66
     millenn
    -0.61
     Gaw
    -0.59
     mortg
    -0.58
     disadvant
    -0.57
     Azerb
    -0.57
     prest
    -0.55
     campaigned
    -0.54
    POSITIVE LOGITS
    ggles
    1.00
    pload
    0.90
    wered
    0.87
    ilet
    0.86
     enable
    0.81
    pless
    0.80
    ppers
    0.79
     RELEASE
    0.78
     provide
    0.77
     learn
    0.77
    Activations Density 0.048%

    No Known Activations