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 24576
    5. 8-RES_FS24576-JB
    6. 21606
    Prev
    Next
    INDEX
    Explanations

    beginnings of sentences with the preposition "at"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    480
    +0.17
    0.8%
    371
    +0.12
    0.6%
    314
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.17
    0.04
    62
    +0.12
    0.02
    499
    +0.09
    0.02
    Negative Logits
    isphere
    -0.62
     grooming
    -0.61
    Ire
    -0.61
    FTWARE
    -0.60
    reditary
    -0.60
    HTTP
    -0.57
    selves
    -0.57
     benefited
    -0.57
    biased
    -0.57
     prescriptions
    -0.56
    POSITIVE LOGITS
    mosp
    1.13
     least
    1.09
    las
    1.02
    hens
    1.00
    yp
    1.00
    onement
    0.96
    rial
    0.89
    abase
    0.87
    roph
    0.85
    ARI
    0.84
    Activations Density 0.063%

    No Known Activations