© 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 24576
    9. 8-RES_FS24576-JB
    11. 19899
    Prev
    Next
    INDEX
    Explanations

    information related to laws and regulations

    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
    59
    +0.09
    0.4%
    763
    +0.09
    0.4%
    745
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    616
    +0.09
    0.03
    100
    +0.09
    0.03
    59
    +0.09
    0.02
    Negative Logits
     distur
    -0.65
     frivol
    -0.62
    aukee
    -0.62
     bout
    -0.62
    ospace
    -0.61
     gor
    -0.61
    76561
    -0.60
    auga
    -0.59
     fastball
    -0.59
     mov
    -0.59
    POSITIVE LOGITS
    neath
    1.41
    lying
    1.21
    graduate
    1.11
    mining
    1.10
    stand
    1.06
    stood
    1.04
    cover
    1.03
    statement
    1.00
    grad
    0.99
    written
    0.98
    Activations Density 0.025%

    No Known Activations