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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1315
    Prev
    Next
    INDEX
    Explanations

    sentences with repeated words or phrases

    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
    203
    +0.11
    0.5%
    393
    +0.11
    0.5%
    152
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.11
    0.07
    713
    +0.11
    0.05
    655
    +0.10
    0.06
    Negative Logits
    uta
    -0.74
    GY
    -0.73
    emi
    -0.72
    ickets
    -0.70
    hered
    -0.69
    asar
    -0.69
    itational
    -0.68
    ut
    -0.66
    ibble
    -0.66
    entials
    -0.65
    POSITIVE LOGITS
     furthermore
    1.27
     moreover
    1.20
     rightfully
    1.19
     consequently
    1.18
     therefore
    1.12
     rightly
    1.07
     secondly
    1.04
     hence
    1.03
     preferably
    1.02
     thus
    1.02
    Activations Density 0.292%

    No Known Activations