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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1757
    Prev
    Next
    INDEX
    Explanations

    headings or sections in a guide or a document

    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
    333
    +0.10
    0.4%
    360
    +0.09
    0.4%
    17
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    395
    +0.10
    0.02
    425
    +0.09
    0.03
    182
    +0.09
    0.02
    Negative Logits
    hist
    -1.02
    comed
    -0.89
    mp
    -0.85
    serv
    -0.85
    sole
    -0.83
    role
    -0.81
    ãĥīãĥ©ãĤ´ãĥ³
    -0.80
    DragonMagazine
    -0.79
    phen
    -0.78
    government
    -0.78
    POSITIVE LOGITS
     XI
    0.97
     VII
    0.94
     2
    0.91
     1
    0.91
     3
    0.90
     4
    0.88
     ½
    0.86
     8
    0.84
     VIII
    0.84
     II
    0.84
    Activations Density 0.058%

    No Known Activations