© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 26520
    Prev
    Next
    INDEX
    Explanations

    concepts related to evolution and creationism debates

    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
    447
    +0.14
    0.6%
    54
    +0.10
    0.5%
    386
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.02
    762
    +0.10
    0.02
    634
    +0.10
    0.02
    Negative Logits
    illes
    -0.75
    stra
    -0.68
     Arrows
    -0.68
    ulla
    -0.66
    LOAD
    -0.66
    thumbnails
    -0.65
    dos
    -0.65
     Polic
    -0.65
    united
    -0.65
    iences
    -0.64
    POSITIVE LOGITS
     debunked
    1.17
    icist
    1.10
     skeptics
    1.05
     denial
    1.03
     scientifically
    0.92
     science
    0.91
     theories
    0.90
     skept
    0.87
     den
    0.86
     hoax
    0.86
    Activations Density 0.048%

    No Known Activations