© 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. 1388
    Prev
    Next
    INDEX
    Explanations

    numerical values and statistics, such as measurements, quantities, or financial figures

    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
    393
    +0.13
    0.6%
    570
    +0.10
    0.4%
    71
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.13
    0.03
    163
    +0.10
    0.04
    357
    +0.09
    0.03
    Negative Logits
    atown
    -0.75
     lest
    -0.74
    xit
    -0.69
    hua
    -0.67
    ridor
    -0.67
    illation
    -0.65
    odore
    -0.65
    arest
    -0.65
    enko
    -0.64
    ADRA
    -0.63
    POSITIVE LOGITS
     configurations
    1.23
     kinds
    1.21
     styles
    1.17
     shapes
    1.16
     scenarios
    1.15
     disciplines
    1.13
     flavors
    1.13
     genres
    1.10
     sorts
    1.10
     types
    1.10
    Activations Density 0.220%

    No Known Activations