© 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. 850
    Prev
    Next
    INDEX
    Explanations

    names and actions related to characters in a narrative or story

    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
    447
    +0.28
    1.3%
    481
    +0.13
    0.6%
    476
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    481
    +0.28
    0.10
    288
    +0.13
    0.10
    399
    +0.10
    0.04
    Negative Logits
     domestically
    -0.69
     proven
    -0.67
     regimes
    -0.66
     advocates
    -0.64
     politically
    -0.64
     Creat
    -0.64
    olitical
    -0.63
     campaigns
    -0.63
     incentives
    -0.63
     historically
    -0.63
    POSITIVE LOGITS
     downstairs
    0.94
     nearby
    0.82
     beck
    0.81
     upstairs
    0.78
     tray
    0.76
     alley
    0.70
     door
    0.69
    stairs
    0.68
    ocular
    0.68
     glim
    0.68
    Activations Density 1.433%

    No Known Activations