© 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. 1483
    Prev
    Next
    INDEX
    Explanations

    information related to military service and political campaigns

    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
    361
    +0.12
    0.5%
    229
    +0.09
    0.4%
    467
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    361
    +0.12
    0.04
    557
    +0.09
    0.04
    422
    +0.09
    0.03
    Negative Logits
    ventus
    -0.72
     Desk
    -0.70
    lihood
    -0.66
     evening
    -0.65
    inki
    -0.64
     immediate
    -0.63
     immediately
    -0.63
     loader
    -0.62
    Elsewhere
    -0.61
     Editors
    -0.61
    POSITIVE LOGITS
    Reviewer
    0.87
    umni
    0.74
     captivity
    0.72
     longest
    0.68
     slavery
    0.67
     culminating
    0.67
    inement
    0.67
    asuring
    0.61
     tort
    0.61
     adolesc
    0.60
    Activations Density 0.973%

    No Known Activations