© 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. 10343
    Prev
    Next
    INDEX
    Explanations

    keywords related to power utilities and electrical grids

    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
    288
    +0.13
    0.6%
    473
    +0.11
    0.5%
    481
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.04
    537
    +0.11
    0.03
    473
    +0.11
    0.03
    Negative Logits
    aked
    -0.89
    orne
    -0.83
    udo
    -0.80
    eworthy
    -0.79
    amia
    -0.78
    oria
    -0.74
    alien
    -0.73
    andon
    -0.73
    eland
    -0.71
    shaw
    -0.71
    POSITIVE LOGITS
     grids
    0.89
     outage
    0.89
     supply
    0.86
     stations
    0.85
     grid
    0.84
     subst
    0.82
     bills
    0.82
     generators
    0.81
     supplying
    0.79
     utilities
    0.78
    Activations Density 0.137%

    No Known Activations