© 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. 9488
    Prev
    Next
    INDEX
    Explanations

    prepositions 'to' and 'at' used 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_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
    718
    +0.12
    0.5%
    608
    +0.12
    0.5%
    49
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.12
    0.07
    49
    +0.12
    0.06
    718
    +0.10
    0.06
    Negative Logits
    calling
    -0.83
    afety
    -0.62
    steam
    -0.61
     Plague
    -0.61
     warranties
    -0.60
     vacancies
    -0.59
     vouchers
    -0.59
     directives
    -0.57
     substitutes
    -0.57
    resy
    -0.56
    POSITIVE LOGITS
     behold
    1.20
    ggles
    1.15
     contemplate
    1.08
    ilet
    0.94
     cherish
    0.93
     achieve
    0.90
     emulate
    0.89
     contend
    0.89
     learn
    0.85
     acquire
    0.85
    Activations Density 0.197%

    No Known Activations