© 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 12288
    9. 8-RES_FS12288-JB
    11. 1605
    Prev
    Next
    INDEX
    Explanations

    words related to validation or legality

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    496
    +0.17
    0.8%
    642
    +0.13
    0.6%
    159
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.17
    0.02
    496
    +0.13
    0.02
    642
    +0.13
    0.01
    Negative Logits
    hedon
    -0.80
    xual
    -0.62
    hani
    -0.62
     Sisters
    -0.60
    hens
    -0.59
    hedral
    -0.59
     darling
    -0.57
     violet
    -0.57
     unsus
    -0.56
     Kut
    -0.56
    POSITIVE LOGITS
    ating
    1.34
    ators
    1.27
    ator
    1.23
    ates
    1.04
    ations
    0.98
    ated
    0.95
    iation
    0.95
    iated
    0.94
    atory
    0.92
    ATING
    0.91
    Activations Density 0.082%

    No Known Activations