© 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. 480
    Prev
    Next
    INDEX
    Explanations

    instances of the word "an" followed by a noun

    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
    572
    +0.12
    0.6%
    523
    +0.11
    0.5%
    381
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.12
    0.10
    362
    +0.11
    0.10
    705
    +0.11
    0.08
    Negative Logits
    stones
    -0.67
     branches
    -0.67
    BLE
    -0.66
    birds
    -0.66
    Topics
    -0.64
    root
    -0.64
    ments
    -0.63
     nests
    -0.62
    hood
    -0.62
     Regions
    -0.62
    POSITIVE LOGITS
    abolic
    1.21
    cients
    1.06
    agram
    1.00
    ointed
    0.98
    oint
    0.95
    alyses
    0.95
    aer
    0.94
    alogue
    0.93
    esthetic
    0.93
    onym
    0.92
    Activations Density 0.291%

    No Known Activations