© 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 12288
    9. 8-RES_FS12288-JB
    11. 9343
    Prev
    Next
    INDEX
    Explanations

    phrases related to natural phenomena and technology, specifically focusing on underground or underwater elements

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    +0.17
    0.8%
    408
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.06
    408
    +0.11
    0.04
    99
    +0.10
    0.03
    Negative Logits
    erion
    -1.10
    bourg
    -1.04
    awed
    -0.92
    rar
    -0.90
    lda
    -0.88
    aido
    -0.87
    aldi
    -0.87
    uid
    -0.83
    abeth
    -0.83
    fred
    -0.82
    POSITIVE LOGITS
     recreation
    0.84
     viewing
    0.76
     photography
    0.74
     skirm
    0.73
     pursuits
    0.73
     fluorescent
    0.72
     robotics
    0.71
     versions
    0.70
     reconnaissance
    0.70
     computing
    0.70
    Activations Density 0.232%

    No Known Activations