© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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 12288
    9. 8-RES_FS12288-JB
    11. 589
    Prev
    Next
    INDEX
    Explanations

    the word "group"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    216
    +0.13
    0.6%
    610
    +0.11
    0.5%
    647
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.13
    0.04
    752
    +0.11
    0.04
    610
    +0.10
    0.03
    Negative Logits
    shire
    -0.65
    Rah
    -0.65
    DAY
    -0.64
     Dame
    -0.63
     Pradesh
    -0.63
    èª
    -0.59
    Bloomberg
    -0.59
    wards
    -0.58
    zzy
    -0.57
    SIGN
    -0.57
    POSITIVE LOGITS
    think
    1.09
    leader
    1.02
     members
    0.98
    members
    0.97
    roups
    0.96
    ings
    0.92
     disbanded
    0.90
    mates
    0.90
    ages
    0.85
     affili
    0.85
    Activations Density 0.101%

    No Known Activations