© 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. 18786
    Prev
    Next
    INDEX
    Explanations

    historical names and entities

    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
    481
    +0.29
    1.4%
    447
    +0.22
    1.1%
    378
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    461
    +0.29
    0.06
    326
    +0.22
    0.07
    481
    +0.14
    0.07
    Negative Logits
    omorphic
    -0.63
    ãĤ¸
    -0.63
    20439
    -0.61
    į
    -0.58
    ãĥīãĥ©ãĤ´ãĥ³
    -0.57
    rss
    -0.57
    Grade
    -0.54
    Bitcoin
    -0.53
    Ü
    -0.53
    OTA
    -0.53
    POSITIVE LOGITS
    ibliography
    0.69
     Appears
    0.56
    fu
    0.56
     Chronicles
    0.54
     lovers
    0.53
    achu
    0.53
    oline
    0.52
    ulent
    0.52
     Disp
    0.52
     Rica
    0.52
    Activations Density 0.546%

    No Known Activations