© 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. 37486
    Prev
    Next
    INDEX
    Explanations

    cautions and warnings related to dangerous situations or harmful substances

    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
    447
    +0.11
    0.5%
    622
    +0.09
    0.4%
    114
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    643
    +0.11
    0.02
    665
    +0.09
    0.03
    669
    +0.08
    0.02
    Negative Logits
     testament
    -0.83
    albeit
    -0.75
    ynthesis
    -0.75
    ciation
    -0.69
    creator
    -0.67
    aez
    -0.65
     creator
    -0.65
    Artist
    -0.64
    ilage
    -0.63
     determination
    -0.63
    POSITIVE LOGITS
     pitfalls
    0.86
    inventoryQuantity
    0.86
     spoilers
    0.84
     stray
    0.81
     overly
    0.76
     EVER
    0.76
     wasting
    0.74
     bother
    0.73
     excessively
    0.73
     excessive
    0.71
    Activations Density 0.051%

    No Known Activations