© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 18529
    Prev
    Next
    INDEX
    Explanations

    mentions of political figures and current events related to government policies

    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
    393
    +0.18
    0.8%
    608
    +0.11
    0.5%
    362
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    605
    +0.18
    0.04
    608
    +0.11
    0.04
    393
    +0.10
    0.04
    Negative Logits
     aloud
    -0.70
    peak
    -0.66
    lins
    -0.66
    hooting
    -0.65
    LEASE
    -0.63
    deals
    -0.62
     accordingly
    -0.62
    duction
    -0.62
    uate
    -0.62
     besides
    -0.61
    POSITIVE LOGITS
     opportunity
    1.21
     same
    1.15
     slightest
    1.10
     ability
    1.08
     utmost
    1.07
     highest
    1.04
     requisite
    1.01
     lowest
    0.99
     widest
    0.99
     greatest
    0.99
    Activations Density 0.086%

    No Known Activations