© 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. 10532
    Prev
    Next
    INDEX
    Explanations

    words related to online content moderation and controversial topics

    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
    189
    +0.10
    0.4%
    635
    +0.09
    0.4%
    496
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.10
    0.09
    73
    +0.09
    0.06
    285
    +0.09
    0.04
    Negative Logits
    Were
    -0.70
    ħĭ
    -0.69
    ãĥĵ
    -0.68
    iece
    -0.66
    vered
    -0.65
    minent
    -0.64
    å¯
    -0.64
    oken
    -0.63
    culus
    -0.63
    estern
    -0.63
    POSITIVE LOGITS
     performs
    1.33
     regulates
    1.33
     promotes
    1.30
     collects
    1.30
     generates
    1.30
     delivers
    1.28
     protects
    1.27
     produces
    1.27
     gives
    1.27
     ignores
    1.27
    Activations Density 0.966%

    No Known Activations