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 49152
    5. 8-RES_FS49152-JB
    6. 23305
    Prev
    Next
    INDEX
    Explanations

    terms and phrases associated with fossil fuels

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    288
    +0.18
    0.8%
    270
    +0.12
    0.5%
    447
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    270
    +0.18
    0.01
    288
    +0.12
    0.01
    258
    +0.11
    0.01
    Negative Logits
    odore
    -0.75
     Niet
    -0.74
    ammad
    -0.72
    èª
    -0.71
     Wong
    -0.71
     CLSID
    -0.70
    yrinth
    -0.70
    ammers
    -0.70
    etheless
    -0.70
     Puzz
    -0.70
    POSITIVE LOGITS
     combustion
    1.04
     extraction
    0.96
     consumption
    0.95
     divest
    0.91
     fuels
    0.90
     reserves
    0.90
     industries
    0.85
     fueled
    0.82
     investments
    0.81
     sources
    0.81
    Activations Density 0.011%

    No Known Activations