© 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 49152
    9. 8-RES_FS49152-JB
    11. 42932
    Prev
    Next
    INDEX
    Explanations

    repetitive phrases involving the word "it."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    46
    +0.10
    0.4%
    314
    +0.10
    0.4%
    336
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.10
    0.02
    699
    +0.10
    0.01
    580
    +0.10
    0.02
    Negative Logits
     srfAttach
    -0.67
    hips
    -0.61
    ggies
    -0.61
     Cards
    -0.60
    arthed
    -0.58
    wcsstore
    -0.58
     advent
    -0.57
     Disabled
    -0.57
    edIn
    -0.56
     arrival
    -0.56
    POSITIVE LOGITS
    alian
    1.20
    atic
    1.05
    self
    0.96
    chy
    0.91
    unes
    0.89
    atically
    0.86
    atics
    0.81
    alia
    0.76
    asca
    0.75
    ucha
    0.72
    Activations Density 0.031%

    No Known Activations