© 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 24576
    9. 8-RES_FS24576-JB
    11. 12246
    Prev
    Next
    INDEX
    Explanations

    government-related text, including mentions of government agencies, applications developed by government agencies, government portals, and administrative processes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    480
    +0.47
    2.4%
    447
    +0.14
    0.7%
    463
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.47
    0.01
    266
    +0.14
    0.08
    566
    +0.11
    0.05
    Negative Logits
    forcer
    -0.59
     ¯
    -0.58
     Darius
    -0.57
    affer
    -0.55
    aurus
    -0.54
     Secondly
    -0.52
    anny
    -0.52
    fourth
    -0.51
     Trailer
    -0.50
     Adin
    -0.50
    POSITIVE LOGITS
     themselves
    0.81
     careers
    0.75
     varying
    0.72
     respective
    0.70
     variants
    0.68
    osponsors
    0.68
     specialize
    0.68
     histories
    0.66
     surn
    0.66
     theirs
    0.64
    Activations Density 0.798%

    No Known Activations