© 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 24576
    9. 8-RES_FS24576-JB
    11. 18752
    Prev
    Next
    INDEX
    Explanations

    numbers related to statistics or measurements

    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
    507
    +0.15
    0.7%
    223
    +0.11
    0.5%
    230
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    507
    +0.15
    0.08
    223
    +0.11
    0.07
    260
    +0.11
    0.07
    Negative Logits
     goodbye
    -0.62
    ador
    -0.61
     Dish
    -0.57
     retreating
    -0.56
    trigger
    -0.55
     grate
    -0.55
     Dos
    -0.54
     histories
    -0.54
     Antarctica
    -0.54
     exiled
    -0.53
    POSITIVE LOGITS
    %.
    1.41
    %,
    1.35
    %;
    1.25
    %-
    1.19
    %
    1.10
     percent
    1.07
    ½
    1.02
    %).
    1.01
    %"
    0.99
    cm
    0.97
    Activations Density 0.180%

    No Known Activations