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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 1536
    9. 8-RES_FS1536-JB
    11. 1107
    Prev
    Next
    INDEX
    Explanations

    action words indicative of causation or production

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    584
    +0.11
    0.5%
    541
    +0.10
    0.5%
    16
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    541
    +0.11
    0.12
    390
    +0.10
    0.08
    217
    +0.10
    0.07
    Negative Logits
    IFT
    -0.66
    Forest
    -0.65
    itte
    -0.63
    ierre
    -0.62
    EEK
    -0.61
    talk
    -0.61
    cot
    -0.61
    oother
    -0.58
    ifted
    -0.57
    POSE
    -0.56
    POSITIVE LOGITS
     by
    1.23
     BY
    1.08
     aback
    0.93
     herein
    0.87
     therein
    0.86
    bys
    0.83
     jointly
    0.81
    by
    0.81
     exclusively
    0.81
     By
    0.81
    Activations Density 1.723%

    No Known Activations