© 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. 26714
    Prev
    Next
    INDEX
    Explanations

    references to specific moments or concepts related to a collective experience

    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
    589
    +0.12
    0.6%
    95
    +0.11
    0.5%
    159
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.12
    0.06
    589
    +0.11
    0.05
    95
    +0.11
    0.04
    Negative Logits
    icons
    -0.84
    immers
    -0.82
    unctions
    -0.81
    acers
    -0.75
    ages
    -0.73
    idity
    -0.72
    features
    -0.71
    mates
    -0.71
    rates
    -0.70
    endants
    -0.69
    POSITIVE LOGITS
     endeavor
    1.03
     predicament
    0.98
     century
    0.93
     week
    0.90
     endeavour
    0.90
     weekend
    0.88
     country
    0.88
     debacle
    0.86
     maze
    0.86
     ordeal
    0.86
    Activations Density 0.082%

    No Known Activations