© 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 12288
    9. 8-RES_FS12288-JB
    11. 870
    Prev
    Next
    INDEX
    Explanations

    words related to the Iraq war, especially specific locations and events

    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
    288
    +0.13
    0.6%
    612
    +0.12
    0.6%
    379
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    612
    +0.13
    0.03
    379
    +0.12
    0.03
    288
    +0.10
    0.03
    Negative Logits
    BOOK
    -0.79
    verett
    -0.76
    box
    -0.76
    boxes
    -0.75
    SHIP
    -0.74
    eries
    -0.71
    ritch
    -0.71
    Premium
    -0.68
    ATURE
    -0.67
    ISTER
    -0.67
    POSITIVE LOGITS
     Hussein
    1.26
     Kurdistan
    1.15
    iyah
    1.06
    Iraq
    0.93
     Iraq
    0.91
     Iraqi
    0.89
    ani
    0.84
    merga
    0.83
    awi
    0.83
    istani
    0.80
    Activations Density 0.077%

    No Known Activations