© 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 12288
    9. 8-RES_FS12288-JB
    11. 10766
    Prev
    Next
    INDEX
    Explanations

    Slavic names

    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
    659
    +0.13
    0.6%
    701
    +0.13
    0.6%
    761
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.03
    701
    +0.13
    0.03
    284
    +0.11
    0.02
    Negative Logits
    kefeller
    -0.94
    iqueness
    -0.82
    acca
    -0.79
    ities
    -0.72
    icum
    -0.67
    ifications
    -0.66
    ifier
    -0.66
    Interstitial
    -0.65
    idental
    -0.65
    iversal
    -0.65
    POSITIVE LOGITS
    hog
    1.08
    orge
    1.05
    rer
    1.00
    rers
    0.92
    ois
    0.85
    ORGE
    0.83
    aux
    0.79
    lla
    0.78
    ttes
    0.78
    ll
    0.77
    Activations Density 0.089%

    No Known Activations