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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1759
    Prev
    Next
    INDEX
    Explanations

    URLs starting with "http."

    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
    373
    +0.11
    0.5%
    383
    +0.11
    0.5%
    439
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.11
    0.04
    439
    +0.11
    0.03
    619
    +0.09
    0.03
    Negative Logits
     Morse
    -0.73
     Labrador
    -0.69
     polyg
    -0.66
    inese
    -0.65
     grav
    -0.62
     elephant
    -0.60
     overshadow
    -0.60
     destro
    -0.60
     succeeding
    -0.58
     moving
    -0.58
    POSITIVE LOGITS
    ://
    2.09
    :/
    1.17
    www
    1.09
    natureconservancy
    1.03
    :\
    1.01
    youtu
    0.95
    docs
    0.95
     www
    0.93
    twitter
    0.87
    books
    0.86
    Activations Density 0.220%

    No Known Activations