© 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. 22219
    Prev
    Next
    INDEX
    Explanations

    conditional phrases that suggest limitations or permissions regarding regulations and policies

    oai_token-act-pair · gpt-4o-miniTriggered 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
    100
    +0.12
    0.6%
    329
    +0.11
    0.5%
    687
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    484
    +0.12
    0.07
    0
    +0.11
    0.06
    393
    +0.11
    0.07
    Negative Logits
    twitch
    -0.78
    ega
    -0.76
    mist
    -0.66
    Morning
    -0.66
    olph
    -0.64
    lash
    -0.63
     trump
    -0.63
     smack
    -0.63
    elo
    -0.61
    stown
    -0.61
    POSITIVE LOGITS
     thereby
    1.08
     related
    0.94
     thereafter
    0.90
     hereafter
    0.90
     thereto
    0.88
     adjoining
    0.86
     hence
    0.85
     consequently
    0.84
     thereof
    0.83
     Related
    0.83
    Activations Density 0.336%

    No Known Activations