© 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 49152
    9. 8-RES_FS49152-JB
    11. 37491
    Prev
    Next
    INDEX
    Explanations

    frequently occurring variations of the "bi-" prefix in botanical or biological terms

    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
    377
    +0.10
    0.5%
    569
    +0.10
    0.4%
    235
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    701
    +0.10
    0.01
    235
    +0.10
    0.01
    220
    +0.10
    0.01
    Negative Logits
    lain
    -0.96
    hips
    -0.71
    mble
    -0.70
    rained
    -0.68
    atility
    -0.66
    eer
    -0.60
    Interstitial
    -0.60
    pter
    -0.60
     anesthesia
    -0.59
     SUP
    -0.59
    POSITIVE LOGITS
    ological
    1.08
    otech
    1.07
    olog
    0.99
    ochem
    0.99
    ographical
    0.98
    ocide
    0.95
    ologists
    0.95
    ologically
    0.95
    omedical
    0.90
    oche
    0.88
    Activations Density 0.006%

    No Known Activations