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. 11599
    Prev
    Next
    INDEX
    Explanations

    descriptions of advanced technology and its implications

    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
    481
    +0.12
    0.6%
    61
    +0.10
    0.5%
    711
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    686
    +0.12
    0.04
    665
    +0.10
    0.06
    541
    +0.10
    0.04
    Negative Logits
    olicy
    -0.68
    rison
    -0.67
    alks
    -0.66
    ECK
    -0.65
     TEAM
    -0.64
    RAY
    -0.63
     ><
    -0.63
    ategory
    -0.61
    OND
    -0.60
    ainers
    -0.59
    POSITIVE LOGITS
    externalActionCode
    0.69
    ãĥ¼ãĥĨ
    0.67
     otherwise
    0.67
     antit
    0.66
    worthy
    0.65
    Rated
    0.64
     incidentally
    0.64
    gru
    0.62
     thereto
    0.62
    mable
    0.62
    Activations Density 0.538%

    No Known Activations