© 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. 630
    Prev
    Next
    INDEX
    Explanations

    references to human life events and experiences

    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
    132
    +0.17
    0.8%
    575
    +0.13
    0.6%
    759
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    132
    +0.17
    0.14
    575
    +0.13
    0.14
    759
    +0.11
    0.10
    Negative Logits
    essee
    -0.88
    sonian
    -0.78
    ery
    -0.78
    mented
    -0.77
     Cosponsors
    -0.77
     shoulders
    -0.71
    PLA
    -0.70
    emark
    -0.70
    oug
    -0.70
    wered
    -0.70
    POSITIVE LOGITS
     expectancy
    1.10
    lihood
    1.04
    blood
    0.93
    guards
    0.85
     vic
    0.84
    guard
    0.84
    liness
    0.81
    mares
    0.78
    scape
    0.78
     everlasting
    0.77
    Activations Density 3.877%

    No Known Activations