© 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. 7748
    Prev
    Next
    INDEX
    Explanations

    references to financial markets and stock indexes, particularly significant declines or losses

    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
    582
    +0.10
    0.4%
    652
    +0.10
    0.4%
    558
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    106
    +0.10
    0.01
    582
    +0.10
    0.01
    577
    +0.09
    0.01
    Negative Logits
    theless
    -0.79
    ãĥĨãĤ£
    -0.70
    bably
    -0.70
    eger
    -0.70
    terday
    -0.69
     GENERAL
    -0.68
    utics
    -0.67
    andom
    -0.67
    ãĤ½
    -0.66
    itably
    -0.66
    POSITIVE LOGITS
    zhou
    1.08
    arters
    1.03
    ars
    1.00
    outs
    0.98
    aroo
    0.91
    ahan
    0.89
    over
    0.89
    overs
    0.89
    ar
    0.89
    bang
    0.87
    Activations Density 0.012%

    No Known Activations