Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 514
    Prev
    Next
    INDEX
    Explanations

    information related to economic activities and social 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
    362
    +0.15
    0.7%
    323
    +0.10
    0.4%
    468
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.15
    0.21
    608
    +0.10
    0.19
    22
    +0.09
    0.14
    Negative Logits
    ãĤ³
    -0.83
    ETH
    -0.81
    URI
    -0.75
    ledge
    -0.74
    anke
    -0.71
    ERROR
    -0.70
    appointed
    -0.70
    00007
    -0.69
    è»
    -0.69
    0000000000000000
    -0.69
    POSITIVE LOGITS
     various
    1.07
     cities
    0.95
     countries
    0.95
     both
    0.94
     workplaces
    0.92
     Europe
    0.90
     rural
    0.87
     behalf
    0.87
     towns
    0.86
     industrialized
    0.86
    Activations Density 3.374%

    No Known Activations