© 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. 22061
    Prev
    Next
    INDEX
    Explanations

    statements of importance or significance regarding acceptable actions or behaviors

    oai_token-act-pair · gpt-4o-miniTriggered 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
    55
    +0.14
    0.6%
    763
    +0.12
    0.5%
    86
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.14
    0.04
    102
    +0.12
    0.04
    541
    +0.10
    0.03
    Negative Logits
    idav
    -0.90
    ertodd
    -0.75
    ifix
    -0.74
    utherland
    -0.72
    restling
    -0.72
    otiation
    -0.69
    cean
    -0.68
     Debor
    -0.67
    ennett
    -0.65
     Aware
    -0.64
    POSITIVE LOGITS
     fair
    0.78
     beneficial
    0.77
     detrimental
    0.70
     essential
    0.70
     unfair
    0.69
     insur
    0.69
     superflu
    0.68
     harmful
    0.67
    selves
    0.67
     tru
    0.67
    Activations Density 0.121%

    No Known Activations