© 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 EvalsExportsinterp-engineNEWAPI 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. 22995
    Prev
    Next
    INDEX
    Explanations

    references to flags and symbols associated with national identity, particularly focusing on the Confederate flag

    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
    359
    +0.12
    0.6%
    481
    +0.12
    0.5%
    667
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.12
    0.03
    614
    +0.12
    0.03
    359
    +0.10
    0.03
    Negative Logits
     latency
    -0.89
    ministic
    -0.82
    cipl
    -0.82
    ynchronous
    -0.80
    bral
    -0.78
    ependent
    -0.77
    humans
    -0.76
    >(
    -0.76
    aido
    -0.73
    advant
    -0.73
    POSITIVE LOGITS
     commemor
    1.39
     adorned
    1.26
     plaque
    1.26
     inscribed
    1.22
     erected
    1.20
     memorial
    1.17
     vandal
    1.16
     mural
    1.15
     decoration
    1.14
     commemorate
    1.13
    Activations Density 0.647%

    No Known Activations