© 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 49152
    9. 8-RES_FS49152-JB
    11. 41493
    Prev
    Next
    INDEX
    Explanations

    gerunds and present participles in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    730
    +0.10
    0.5%
    453
    +0.10
    0.5%
    671
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    453
    +0.10
    0.03
    408
    +0.10
    0.03
    219
    +0.09
    0.02
    Negative Logits
    yip
    -0.72
    Ô
    -0.71
    putable
    -0.71
    glas
    -0.69
    respect
    -0.69
     subur
    -0.67
    available
    -0.67
    alde
    -0.66
    redited
    -0.65
    ickr
    -0.65
    POSITIVE LOGITS
    ly
    1.49
    LY
    1.23
    ingly
    0.99
     spectacle
    0.92
     ordeal
    0.84
     tale
    0.83
    ously
    0.74
     ord
    0.71
     visuals
    0.71
    iously
    0.71
    Activations Density 0.048%

    No Known Activations