© 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. 1968
    Prev
    Next
    INDEX
    Explanations

    financial amounts represented in dollars

    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
    665
    +0.12
    0.5%
    314
    +0.11
    0.5%
    685
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    685
    +0.12
    0.03
    680
    +0.11
    0.03
    310
    +0.10
    0.04
    Negative Logits
     Epidem
    -0.65
    ullivan
    -0.61
     VIDEOS
    -0.59
    ardy
    -0.59
     brunt
    -0.58
     Statistics
    -0.57
    itionally
    -0.56
     Gallup
    -0.56
     Administ
    -0.56
     Typ
    -0.56
    POSITIVE LOGITS
    $$$$
    0.88
    .$
    0.85
     dollars
    0.83
    -$
    0.83
     euros
    0.82
     apiece
    0.82
     bucks
    0.82
    USD
    0.82
    $.
    0.81
     sake
    0.81
    Activations Density 0.157%

    No Known Activations