Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 529
    Prev
    Next
    INDEX
    Explanations

    phrases related to governmental processes and procedures

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    460
    +0.13
    0.6%
    312
    +0.13
    0.6%
    762
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.13
    0.15
    132
    +0.13
    0.13
    762
    +0.11
    0.17
    Negative Logits
    ufact
    -0.71
    lar
    -0.64
    hao
    -0.62
    resa
    -0.61
    ammy
    -0.61
    pat
    -0.60
    anamo
    -0.57
    algia
    -0.56
     TC
    -0.56
     tomat
    -0.56
    POSITIVE LOGITS
     whatsoever
    0.79
    .
    0.76
     ILCS
    0.73
    Footnote
    0.72
     glance
    0.71
    ,
    0.69
     thereafter
    0.68
    ;
    0.67
     resembling
    0.65
    .(
    0.65
    Activations Density 3.105%

    No Known Activations