© 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. 887
    Prev
    Next
    INDEX
    Explanations

    comparisons or emphasis on quantities or levels

    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
    655
    +0.14
    0.7%
    373
    +0.12
    0.6%
    718
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    655
    +0.14
    0.07
    724
    +0.12
    0.05
    183
    +0.11
    0.05
    Negative Logits
     largeDownload
    -0.73
    NEY
    -0.68
    20439
    -0.66
    ffen
    -0.65
    OSP
    -0.63
     derog
    -0.62
    ilic
    -0.62
    EMENT
    -0.62
    oris
    -0.61
    CLASSIFIED
    -0.60
    POSITIVE LOGITS
    whelming
    0.90
     tripled
    0.89
     doubled
    0.83
     doubling
    0.80
    represent
    0.73
    represented
    0.71
     1500
    0.71
     8000
    0.71
    atos
    0.70
     half
    0.70
    Activations Density 0.094%

    No Known Activations