© 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. 6882
    Prev
    Next
    INDEX
    Explanations

    information related to terrorism and political conflicts

    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
    480
    +0.49
    2.6%
    393
    +0.11
    0.6%
    266
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.49
    0.14
    575
    +0.11
    0.10
    266
    +0.11
    0.10
    Negative Logits
    minecraft
    -0.63
     Helpful
    -0.61
    htaking
    -0.58
    ï¸ı
    -0.54
    Pg
    -0.54
     Greenberg
    -0.53
    rawdownloadcloneembedreportprint
    -0.53
    udic
    -0.52
    20439
    -0.52
    POR
    -0.51
    POSITIVE LOGITS
    rique
    0.64
     equivalents
    0.56
    eers
    0.51
     respectively
    0.51
     powerhouse
    0.50
     supremacy
    0.49
     elsewhere
    0.49
     itself
    0.49
     alike
    0.48
     masters
    0.47
    Activations Density 1.205%

    No Known Activations