© 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. 10332
    Prev
    Next
    INDEX
    Explanations

    common phrases or structures in a narrative, such as descriptions of travel and experiences

    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
    214
    +0.11
    0.5%
    581
    +0.11
    0.5%
    619
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    581
    +0.11
    0.02
    619
    +0.11
    0.03
    158
    +0.10
    0.03
    Negative Logits
    qus
    -0.68
    WM
    -0.66
     abound
    -0.65
    rers
    -0.64
    dm
    -0.63
     Bron
    -0.63
    haven
    -0.63
     Curt
    -0.62
    paren
    -0.62
    Ws
    -0.60
    POSITIVE LOGITS
     individually
    1.54
     separately
    1.21
     unique
    1.03
     independently
    1.01
     varying
    0.99
     distinct
    0.97
     differing
    0.96
     differently
    0.88
     uniquely
    0.87
     respective
    0.87
    Activations Density 0.219%

    No Known Activations