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. 21395
    Prev
    Next
    INDEX
    Explanations

    references to specific technical products or technologies such as a specific brand of CPU

    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
    86
    +0.13
    0.6%
    420
    +0.11
    0.5%
    496
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.01
    208
    +0.11
    0.01
    547
    +0.10
    0.01
    Negative Logits
    assetsadobe
    -0.68
    enegger
    -0.67
     Curve
    -0.67
    chrom
    -0.66
    BuyableInstoreAndOnline
    -0.66
    pmwiki
    -0.65
    iferation
    -0.65
    oho
    -0.64
    lime
    -0.63
    Americans
    -0.63
    POSITIVE LOGITS
    lyn
    1.12
    rette
    1.11
     Valkyrie
    0.94
    lla
    0.89
    ilon
    0.86
    lling
    0.83
    llers
    0.82
    lynn
    0.81
    eve
    0.76
     Eve
    0.76
    Activations Density 0.028%

    No Known Activations