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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1635
    Prev
    Next
    INDEX
    Explanations

    mentions of sports teams

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    167
    +0.11
    0.5%
    302
    +0.11
    0.5%
    339
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    652
    +0.11
    0.06
    247
    +0.11
    0.05
    31
    +0.11
    0.06
    Negative Logits
    tml
    -0.86
     drip
    -0.72
    tain
    -0.70
    forward
    -0.69
    dit
    -0.68
    ibrary
    -0.68
    fect
    -0.67
    alam
    -0.66
    ously
    -0.66
    pire
    -0.65
    POSITIVE LOGITS
    mates
    1.19
     mates
    1.14
    members
    1.09
    rons
    0.95
    liquid
    0.91
    mate
    0.89
    peak
    0.88
     members
    0.88
    sters
    0.85
     motto
    0.84
    Activations Density 0.447%

    No Known Activations