© 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. 26692
    Prev
    Next
    INDEX
    Explanations

    instances of the word "lur" and its variations

    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
    742
    +0.13
    0.6%
    99
    +0.10
    0.5%
    326
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    742
    +0.13
    0.01
    326
    +0.10
    0.01
    99
    +0.10
    0.01
    Negative Logits
     Errors
    -0.71
    enegger
    -0.70
     Fathers
    -0.70
     guiActiveUnfocused
    -0.66
    ortium
    -0.66
     Merchants
    -0.65
     Pwr
    -0.65
     Breaker
    -0.64
     Luther
    -0.63
    ELS
    -0.63
    POSITIVE LOGITS
    uten
    0.94
    ker
    0.88
    ping
    0.87
    iously
    0.86
    aces
    0.82
    ormal
    0.82
    ace
    0.80
    kers
    0.79
    ge
    0.79
    ous
    0.78
    Activations Density 0.008%

    No Known Activations