© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 43094
    Prev
    Next
    INDEX
    Explanations

    discussions regarding anonymity and the sharing of sensitive information

    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
    412
    +0.10
    0.5%
    425
    +0.10
    0.4%
    91
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    760
    +0.10
    0.04
    679
    +0.10
    0.05
    425
    +0.09
    0.04
    Negative Logits
     fundamentals
    -0.67
     pedest
    -0.64
     sights
    -0.63
    tumblr
    -0.61
    ãĢIJ
    -0.60
    joice
    -0.58
     Confederate
    -0.58
     Avenger
    -0.58
     wre
    -0.58
     Asgard
    -0.57
    POSITIVE LOGITS
     anonymity
    1.15
     surn
    0.91
     confidentiality
    0.83
     anonym
    0.81
     emailed
    0.81
    ispers
    0.78
     disclosed
    0.77
    otaur
    0.77
     anonymously
    0.76
     interviewed
    0.76
    Activations Density 0.272%

    No Known Activations