© 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. 18494
    Prev
    Next
    INDEX
    Explanations

    locations and organizations

    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
    87
    +0.25
    1.3%
    620
    +0.10
    0.5%
    235
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.25
    0.11
    575
    +0.10
    0.08
    537
    +0.09
    0.07
    Negative Logits
    ITNESS
    -0.69
     Canaver
    -0.62
    itialized
    -0.52
    )=(
    -0.51
    ettel
    -0.50
    ATURE
    -0.49
    Journal
    -0.47
    urse
    -0.46
    ativity
    -0.45
    GoldMagikarp
    -0.45
    POSITIVE LOGITS
     and
    0.92
     or
    0.78
     etc
    0.76
    +.
    0.71
     et
    0.68
     &
    0.67
    ®,
    0.65
     AND
    0.64
    /.
    0.63
    *.
    0.63
    Activations Density 0.902%

    No Known Activations