© 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. 5580
    Prev
    Next
    INDEX
    Explanations

    phrases related to competitive sports and locker room scenarios

    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
    +0.27
    1.3%
    351
    +0.12
    0.6%
    517
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.27
    0.07
    43
    +0.12
    0.06
    546
    +0.11
    0.07
    Negative Logits
    atorium
    -0.82
    ":"/
    -0.65
    ¬¼
    -0.64
    orem
    -0.62
    iple
    -0.62
    ãĤ¦ãĤ¹
    -0.62
    emaker
    -0.62
    rament
    -0.61
    izon
    -0.61
    forth
    -0.61
    POSITIVE LOGITS
     etc
    1.10
     whereas
    1.07
     namely
    0.92
    etc
    0.85
     respectively
    0.85
     haha
    0.85
     aka
    0.85
     huh
    0.84
     preferring
    0.83
     including
    0.83
    Activations Density 0.347%

    No Known Activations