© 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 12288
    9. 8-RES_FS12288-JB
    11. 799
    Prev
    Next
    INDEX
    Explanations

    references to specific athlete-related terms and sports-related contexts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    481
    +0.17
    0.8%
    422
    +0.10
    0.5%
    564
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    422
    +0.17
    0.04
    564
    +0.10
    0.05
    288
    +0.10
    0.05
    Negative Logits
    ortium
    -0.78
    oras
    -0.74
    laus
    -0.71
    IRD
    -0.62
     warr
    -0.62
    oris
    -0.59
     Telesc
    -0.59
    Operation
    -0.58
     href
    -0.58
    anson
    -0.57
    POSITIVE LOGITS
     podium
    0.91
    plays
    0.83
    waves
    0.81
     floor
    0.80
     sidelines
    0.78
     aisle
    0.77
    walk
    0.75
     antics
    0.74
     scoreboard
    0.70
     brawl
    0.70
    Activations Density 0.220%

    No Known Activations