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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 6987
    Prev
    Next
    INDEX
    Explanations

    mentions of presence or being present in various contexts

    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
    447
    +0.12
    0.6%
    504
    +0.10
    0.5%
    327
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.12
    0.04
    168
    +0.10
    0.03
    517
    +0.10
    0.03
    Negative Logits
     subp
    -0.75
     reliance
    -0.71
     spir
    -0.69
     downward
    -0.67
     dens
    -0.66
     Flavoring
    -0.65
     regression
    -0.65
     fisheries
    -0.64
     downwards
    -0.64
    division
    -0.64
    POSITIVE LOGITS
     cheering
    0.76
    ACTION
    0.73
     booth
    0.73
     Guests
    0.71
     Saturday
    0.71
     during
    0.70
    Saturday
    0.70
    aturday
    0.70
     superv
    0.70
    atel
    0.69
    Activations Density 0.172%

    No Known Activations