© 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 768
    9. 8-RES_FS768-JB
    11. 767
    Prev
    Next
    INDEX
    Explanations

    countries or locations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    481
    +0.22
    1.0%
    447
    +0.19
    0.9%
    67
    +0.16
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    67
    +0.22
    0.20
    57
    +0.19
    0.14
    288
    +0.16
    0.18
    Negative Logits
    inventoryQuantity
    -0.67
    colo
    -0.65
    ynt
    -0.64
    lihood
    -0.64
    atin
    -0.63
    TAG
    -0.63
    onym
    -0.63
    naissance
    -0.62
    LEASE
    -0.62
    rious
    -0.61
    POSITIVE LOGITS
     Airlines
    1.00
    's
    0.84
     Arabia
    0.84
     Springs
    0.81
     Yard
    0.79
    iens
    0.76
     Skies
    0.76
     Highlands
    0.75
    ESE
    0.75
     Airways
    0.74
    Activations Density 2.406%

    No Known Activations