© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 24225
    Prev
    Next
    INDEX
    Explanations

    statements of disbelief or emphasis

    oai_token-act-pair · gpt-4o-miniTriggered 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
    54
    +0.11
    0.5%
    423
    +0.11
    0.5%
    241
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    45
    +0.11
    0.03
    760
    +0.11
    0.03
    151
    +0.10
    0.04
    Negative Logits
    tnc
    -0.77
    lain
    -0.73
    ousel
    -0.72
    ription
    -0.69
    aciously
    -0.68
    oulder
    -0.67
    umbn
    -0.67
    oise
    -0.66
    ullah
    -0.66
    uments
    -0.65
    POSITIVE LOGITS
     kidding
    0.86
     speaking
    0.85
    entimes
    0.83
     tho
    0.78
     congr
    0.75
     Speaking
    0.75
     congratulations
    0.73
     clothed
    0.73
     importantly
    0.70
    !?
    0.69
    Activations Density 0.117%

    No Known Activations