© 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. 22258
    Prev
    Next
    INDEX
    Explanations

    references to topics of discussion or discourse

    oai_token-act-pair · gpt-4o-miniTriggered 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
    481
    +0.24
    1.1%
    447
    +0.16
    0.7%
    648
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    557
    +0.24
    0.04
    755
    +0.16
    0.04
    648
    +0.09
    0.04
    Negative Logits
    TY
    -0.67
    OFF
    -0.66
    ENSE
    -0.63
    NT
    -0.63
     Sabha
    -0.61
    ometer
    -0.61
    CBC
    -0.60
    ND
    -0.60
    rained
    -0.60
    fax
    -0.60
    POSITIVE LOGITS
     topics
    0.85
     behalf
    0.80
     cryptocurrencies
    0.77
     Causes
    0.76
     Principles
    0.73
     UFOs
    0.72
     genetics
    0.72
     Transformation
    0.69
     Understanding
    0.69
     Surv
    0.68
    Activations Density 0.221%

    No Known Activations