© 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 49152
    9. 8-RES_FS49152-JB
    11. 26533
    Prev
    Next
    INDEX
    Explanations

    the name "Derek" in various contexts

    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
    549
    +0.11
    0.5%
    195
    +0.10
    0.5%
    326
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.01
    81
    +0.10
    0.01
    326
    +0.10
    0.01
    Negative Logits
    izations
    -0.94
    hips
    -0.89
    ILA
    -0.77
    ropolitan
    -0.76
    folk
    -0.71
    isations
    -0.71
    places
    -0.71
    fully
    -0.70
    heast
    -0.69
    Īè
    -0.69
    POSITIVE LOGITS
     Carr
    0.92
     Barker
    0.76
     Lawson
    0.76
     Barnett
    0.75
     Smart
    0.74
     Prescott
    0.74
     McKenzie
    0.74
     Derek
    0.73
     Dunn
    0.72
     Crosby
    0.71
    Activations Density 0.004%

    No Known Activations