© 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. 45698
    Prev
    Next
    INDEX
    Explanations

    mentions of Conor McGregor in a variety of contexts, particularly regarding fights

    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
    +0.20
    0.9%
    288
    +0.15
    0.7%
    481
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    502
    +0.20
    0.01
    46
    +0.15
    0.01
    224
    +0.14
    0.01
    Negative Logits
    PsyNetMessage
    -0.76
    ãĤ·ãĥ£
    -0.74
    cle
    -0.72
    ktop
    -0.72
    clerosis
    -0.71
    semble
    -0.69
    DCS
    -0.69
    DonaldTrump
    -0.68
    anke
    -0.66
    krit
    -0.64
    POSITIVE LOGITS
    itton
    0.70
    STON
    0.65
    orious
    0.65
     McGregor
    0.64
     Mayweather
    0.63
    agher
    0.63
    donald
    0.62
    hetti
    0.60
     selfies
    0.60
     Yard
    0.59
    Activations Density 0.008%

    No Known Activations