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 3072
    5. 8-RES_FS3072-JB
    6. 2196
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial data, statistics, or technical specifications

    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
    649
    +0.14
    0.6%
    20
    +0.11
    0.5%
    445
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    699
    +0.14
    0.04
    194
    +0.11
    0.04
    260
    +0.11
    0.05
    Negative Logits
    ĪĴ
    -0.79
    stones
    -0.74
    hov
    -0.72
    heid
    -0.71
    step
    -0.69
    zos
    -0.68
    ©¶æ
    -0.66
    pher
    -0.65
    pell
    -0.64
    hops
    -0.64
    POSITIVE LOGITS
    itarian
    1.04
     length
    0.93
     amount
    0.83
     lifespan
    0.81
     likelihood
    0.77
     number
    0.76
     Eclipse
    0.75
     package
    0.72
     annual
    0.72
     height
    0.72
    Activations Density 0.608%

    No Known Activations