© 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. 4515
    Prev
    Next
    INDEX
    Explanations

    information related to legal cases, court proceedings, and political events

    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
    480
    +0.18
    0.8%
    20
    +0.09
    0.4%
    271
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.18
    0.06
    393
    +0.09
    0.08
    473
    +0.09
    0.04
    Negative Logits
    Used
    -0.65
    à¨
    -0.58
     dehuman
    -0.58
     invention
    -0.57
    interstitial
    -0.57
     systematic
    -0.56
     discriminated
    -0.56
     misogyn
    -0.56
     vulgar
    -0.55
     pearl
    -0.54
    POSITIVE LOGITS
     likely
    0.95
    likely
    0.95
     2019
    0.94
     will
    0.93
     unlikely
    0.89
     meantime
    0.88
    will
    0.87
     2020
    0.87
     2021
    0.86
    assuming
    0.86
    Activations Density 0.806%

    No Known Activations