© 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. 41508
    Prev
    Next
    INDEX
    Explanations

    sequences or mentions of structured processes and rankings

    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
    242
    +0.11
    0.5%
    307
    +0.10
    0.4%
    390
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.11
    0.02
    62
    +0.10
    0.01
    4
    +0.10
    0.01
    Negative Logits
    tek
    -0.94
    tu
    -0.71
    tty
    -0.67
    td
    -0.66
    gets
    -0.65
    tn
    -0.64
    oubted
    -0.64
     Nadu
    -0.63
    immer
    -0.62
    hawk
    -0.62
    POSITIVE LOGITS
     alphabet
    1.00
    lies
    0.91
    liness
    0.89
     chronological
    0.85
     sorted
    0.79
     comma
    0.79
    phabet
    0.78
     sorting
    0.69
     dictated
    0.68
     rounding
    0.67
    Activations Density 0.016%

    No Known Activations