© 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 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. 10322
    Prev
    Next
    INDEX
    Explanations

    proper names and references to individuals

    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
    470
    +0.12
    0.5%
    585
    +0.12
    0.5%
    304
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    470
    +0.12
    0.02
    585
    +0.12
    0.02
    304
    +0.10
    0.02
    Negative Logits
     coma
    -0.67
     optionally
    -0.67
     heroin
    -0.66
    ĪĴ
    -0.64
    romeda
    -0.63
     snag
    -0.63
     Cortex
    -0.63
     MIA
    -0.62
    axter
    -0.62
     morphine
    -0.62
    POSITIVE LOGITS
    enthal
    0.93
    igans
    0.87
    enger
    0.85
    know
    0.81
    bugs
    0.81
    mann
    0.80
    fram
    0.78
    ner
    0.77
    nder
    0.77
    heit
    0.77
    Activations Density 0.029%

    No Known Activations