© 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 EvalsExportsinterp-engineNEWAPI 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. 2390
    Prev
    Next
    INDEX
    Explanations

    phrases related to addressing challenges or problems

    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
    447
    +0.14
    0.7%
    622
    +0.12
    0.6%
    455
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    622
    +0.14
    0.02
    665
    +0.12
    0.02
    468
    +0.11
    0.02
    Negative Logits
    abet
    -0.83
    RIP
    -0.78
    AMY
    -0.69
    appointed
    -0.67
    iew
    -0.65
    estate
    -0.64
    akin
    -0.64
    anwhile
    -0.63
    uary
    -0.63
    %%%%
    -0.61
    POSITIVE LOGITS
     tackle
    0.79
     tricky
    0.78
     thorn
    0.76
     harder
    0.74
     tackling
    0.73
     hurdles
    0.69
     challenging
    0.68
    itely
    0.67
    illes
    0.66
     uphill
    0.66
    Activations Density 0.055%

    No Known Activations