© 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 3072
    9. 8-RES_FS3072-JB
    11. 750
    Prev
    Next
    INDEX
    Explanations

    mentions of the action "pay" with varying degrees of emphasis

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    366
    +0.10
    0.4%
    552
    +0.09
    0.4%
    59
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    366
    +0.10
    0.04
    552
    +0.09
    0.05
    688
    +0.09
    0.05
    Negative Logits
    ierre
    -0.85
    bum
    -0.70
    âĸ¬
    -0.63
    OPLE
    -0.62
    Aim
    -0.62
    KEN
    -0.61
    -----
    -0.61
     Moder
    -0.61
    ------
    -0.60
    Í
    -0.60
    POSITIVE LOGITS
    checks
    1.11
     dues
    1.09
     dividends
    1.05
     homage
    1.05
     attention
    1.00
     tribute
    0.98
     taxes
    0.97
     royalties
    0.95
     dearly
    0.94
     heed
    0.93
    Activations Density 0.404%

    No Known Activations