© 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. 10617
    Prev
    Next
    INDEX
    Explanations

    phrases related to interviews and discussions, particularly about careers, struggles, and support

    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
    638
    +0.09
    0.4%
    88
    +0.09
    0.4%
    681
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    638
    +0.09
    0.03
    632
    +0.09
    0.03
    92
    +0.09
    0.03
    Negative Logits
    *)
    -0.73
    clock
    -0.68
    dl
    -0.67
     largeDownload
    -0.65
     Persons
    -0.64
    blers
    -0.64
    iciency
    -0.64
    ivable
    -0.62
    Grade
    -0.61
    bris
    -0.61
    POSITIVE LOGITS
     misconceptions
    1.03
     challeng
    0.88
     inspir
    0.83
     reluct
    0.82
     mindset
    0.79
     philosophies
    0.78
     overcoming
    0.77
     filmmaking
    0.76
     philosophy
    0.74
     redes
    0.72
    Activations Density 0.362%

    No Known Activations