© 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 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. 10122
    Prev
    Next
    INDEX
    Explanations

    descriptions of events or situations that are considered important or significant

    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
    +0.28
    1.4%
    680
    +0.10
    0.5%
    756
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.28
    -0.03
    750
    +0.10
    0.06
    608
    +0.09
    0.09
    Negative Logits
    quartered
    -0.66
    interstitial
    -0.64
    ilus
    -0.64
    othermal
    -0.60
    currently
    -0.58
    anges
    -0.56
    onga
    -0.56
    tis
    -0.55
    ciplinary
    -0.55
    ilial
    -0.54
    POSITIVE LOGITS
     yesterday
    1.05
     earlier
    0.94
     last
    0.94
    terday
    0.83
     originally
    0.80
     beforehand
    0.79
     LAST
    0.69
     previous
    0.67
     initially
    0.66
     prior
    0.66
    Activations Density 1.236%

    No Known Activations