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. 201
    Prev
    Next
    INDEX
    Explanations

    sections related to various topics such as entertainment, technology, sports, and cultural issues

    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
    447
    +0.16
    0.7%
    481
    +0.12
    0.5%
    608
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.16
    0.18
    8
    +0.12
    0.11
    562
    +0.09
    0.09
    Negative Logits
    orum
    -0.67
    hene
    -0.64
     equals
    -0.63
    iken
    -0.60
    Copyright
    -0.58
    sis
    -0.58
    perfect
    -0.57
    he
    -0.57
     shouldn
    -0.56
     hadn
    -0.56
    POSITIVE LOGITS
    interstitial
    0.80
    senal
    0.77
     another
    0.76
    cellaneous
    0.73
     other
    0.70
    another
    0.68
     Tatt
    0.64
     neighbouring
    0.64
    ammy
    0.62
     newcom
    0.62
    Activations Density 5.967%

    No Known Activations