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

    Neuronpedia

    APICircuit 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. 1342
    Prev
    Next
    INDEX
    Explanations

    words related to power and energy

    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
    755
    +0.13
    0.6%
    288
    +0.10
    0.5%
    634
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.11
    523
    +0.10
    0.07
    755
    +0.10
    0.06
    Negative Logits
     Von
    -0.77
    algia
    -0.74
    eday
    -0.72
    eryl
    -0.72
     Bei
    -0.69
    ĸļ
    -0.68
    auder
    -0.67
    romeda
    -0.66
    oslov
    -0.63
    owitz
    -0.63
    POSITIVE LOGITS
    houses
    1.11
    plant
    0.93
     outage
    0.93
    stroke
    0.93
    lifting
    0.88
     grids
    0.86
    train
    0.85
     generators
    0.83
     cords
    0.82
    assisted
    0.82
    Activations Density 1.791%

    No Known Activations