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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 15949
    Prev
    Next
    INDEX
    Explanations

    text related to religious teachings and forgiveness

    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
    480
    +0.39
    1.9%
    447
    +0.16
    0.8%
    87
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.39
    0.03
    266
    +0.16
    0.10
    397
    +0.13
    0.06
    Negative Logits
     frontline
    -0.65
     Spotlight
    -0.65
     Cassidy
    -0.64
    Focus
    -0.63
     pipeline
    -0.62
     OPS
    -0.61
     Trend
    -0.60
    TPS
    -0.60
     Panda
    -0.60
    EPA
    -0.59
    POSITIVE LOGITS
     hath
    1.17
     thence
    1.10
    ,—
    1.06
     unto
    1.04
     ye
    0.98
     whence
    0.97
     thou
    0.97
     shall
    0.91
     negro
    0.91
     afore
    0.84
    Activations Density 1.173%

    No Known Activations