© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9372
    Prev
    Next
    INDEX
    Explanations

    references to political actions and international relationships with a focus on events involving the United States

    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
    608
    +0.13
    0.6%
    290
    +0.11
    0.5%
    299
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.04
    299
    +0.11
    0.03
    259
    +0.10
    0.03
    Negative Logits
    standing
    -0.74
     ticking
    -0.68
    etooth
    -0.65
    iction
    -0.65
    rated
    -0.64
    paced
    -0.63
     certs
    -0.63
    arius
    -0.62
    iveness
    -0.62
    ratulations
    -0.61
    POSITIVE LOGITS
     Disneyland
    0.92
     Thailand
    0.92
     China
    0.92
     Mongolia
    0.91
     Ethiopia
    0.89
     Uganda
    0.87
     Africa
    0.87
     India
    0.87
     meet
    0.87
     Somalia
    0.86
    Activations Density 0.105%

    No Known Activations