© 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 24576
    9. 8-RES_FS24576-JB
    11. 18889
    Prev
    Next
    INDEX
    Explanations

    phrases related to sports events or game statistics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    218
    +0.14
    0.6%
    24
    +0.09
    0.4%
    51
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.14
    0.02
    608
    +0.09
    0.03
    24
    +0.09
    0.02
    Negative Logits
    ensical
    -0.72
    hari
    -0.70
    gnu
    -0.69
     contin
    -0.69
    tle
    -0.67
     condol
    -0.67
    elling
    -0.66
    cart
    -0.66
    lat
    -0.66
    reported
    -0.65
    POSITIVE LOGITS
     halftime
    1.04
     adulthood
    1.04
     season
    0.94
     retirement
    0.91
     Thanksgiving
    0.86
     offseason
    0.85
     2017
    0.84
    imester
    0.83
     Week
    0.81
     adolescence
    0.81
    Activations Density 0.084%

    No Known Activations