© 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. 18356
    Prev
    Next
    INDEX
    Explanations

    references to specific military vessels, specifically aircraft carriers with the prefix "USS" followed by their name

    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
    314
    +0.11
    0.5%
    59
    +0.10
    0.5%
    683
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    358
    +0.11
    0.01
    210
    +0.10
    0.01
    538
    +0.10
    0.01
    Negative Logits
    rency
    -0.74
    izophren
    -0.72
     exerc
    -0.71
     pandemonium
    -0.71
    ously
    -0.70
     stroke
    -0.67
     shaping
    -0.65
    xus
    -0.65
    aum
    -0.65
    zzle
    -0.65
    POSITIVE LOGITS
     Voyager
    0.92
     Enterprise
    0.91
    USS
    0.88
     USS
    0.85
     Titanic
    0.83
    OC
    0.77
     Yamato
    0.76
    GE
    0.76
     Fitzgerald
    0.75
    arrison
    0.74
    Activations Density 0.023%

    No Known Activations