© 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. 10193
    Prev
    Next
    INDEX
    Explanations

    time-related phrases, such as durations or periods

    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
    73
    +0.11
    0.5%
    680
    +0.10
    0.5%
    314
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.11
    0.05
    680
    +0.10
    0.05
    258
    +0.10
    0.06
    Negative Logits
    inki
    -0.85
    ansk
    -0.67
    ioned
    -0.67
    ucker
    -0.62
    dra
    -0.60
     flakes
    -0.57
    ater
    -0.57
    iru
    -0.56
    oya
    -0.56
     Finish
    -0.54
    POSITIVE LOGITS
     decades
    1.07
     awhile
    1.05
     lately
    1.04
     millennia
    1.00
     years
    0.95
     YEAR
    0.95
    since
    0.95
     since
    0.89
     centuries
    0.87
    months
    0.84
    Activations Density 0.423%

    No Known Activations