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. 11705
    Prev
    Next
    INDEX
    Explanations

    references to legal matters and legal terminology

    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
    253
    +0.11
    0.5%
    302
    +0.10
    0.5%
    328
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    328
    +0.11
    0.01
    666
    +0.10
    0.01
    711
    +0.10
    0.01
    Negative Logits
    earchers
    -0.88
    kamp
    -0.75
    igslist
    -0.72
    adr
    -0.70
    romeda
    -0.69
    atu
    -0.68
    seekers
    -0.67
    reens
    -0.66
    chers
    -0.65
    arching
    -0.65
    POSITIVE LOGITS
     inconven
    1.01
     tweaks
    0.97
    itized
    0.96
     inconvenience
    0.95
     annoyance
    0.90
     annoy
    0.87
     adjustments
    0.86
     league
    0.85
     leagues
    0.85
    ities
    0.80
    Activations Density 0.018%

    No Known Activations